Starting with your form intersections, overall you are showing progress here (the second page definitely shows a lot more comfort with intersecting curved surfaces than the first page), although the first page is admittedly a little difficult to parse, for a couple reasons. Firstly, you seem to have a lot more additional lines present on the page, which make it hard to interpret which one's meant to be the intersection. In some cases (like around 8 on the first page) you attempt to correct some mistakes, which further adds confusion, and in others (like 7) there's more than just the visible intersection being drawn - you might be trying to define the intersection all the way around. In general, try to focus only on the part of the intersection that will be visible. Also, there are some cases where your cylinders' orientations are kind of hard to make out - filling the end facing the viewer with some hatching would definitely help in that regard.

Your second page is definitely more clear - part of that is the use of a different colour (I'll touch on this in a second), but it really comes down more to the fact that you're focusing more on the visible intersection (though you seem to be drawing the invisible parts in some cases).

Now, while using a different colour for your intersections does help them become more clear for me, they also have a detrimental effect to the exercise, giving the impression that the intersection lines are separate from the drawing itself. In fact, they're not - the intersection lines themselves are actually a concrete part of these forms - they define the relationships between them, and behave similarly to "weld lines" when you've got objects actually being welded together. So, drawing them in black along with the rest of the exercise can help make the connections in our brain of what these lines are physically meant to represent in the world. While this can make things a bit harder to make out, as long as you avoid drawing anything else (drawing unseen portions of the intersections, or other additional lines), it should be okay.

While you are indeed improving as you move through these pages, I have marked out some corrections on the third page for anything that stood out as being especially incorrect. When it comes to figuring out the intersections between curving surfaces (which is notoriously difficult), I like to focus on first figuring out the specific way in which the surfaces in question curve. A sphere, for instance, curves in every possible direction, but depending on the form you're intersecting with, only one direction of curvature will actually be relevant. For example, if you intersect a box with a sphere, the specific curvatures of the sphere that are relevant will be those that align to the box's own faces. So for example, here we've got a box and a ball intersecting. Two of the box's faces actually touches the ball, and we can actually look at their intersections one at a time. With the box's top plane, we see that only the blue curve along the ball's surface is relevant, so this dictates how that intersection occurs. For the box's front plane, the green curvature is relevant, and so that's what determines its intersection with the ball. And of course, both of these intersection lines must touch at the edge between them, which is when we jump from following one direction of curvature to the other.

Applying this to two separate rounded forms however is considerably more difficult, but we can use the same principles. As shown here, there's only one direction in which the cylinder curves (the green arrows), and it's the orientation of the cylinder itself that determines which curvature of the sphere's surface we need to be looking at (the blue arrow). From there, we need to figure out how to go from the cylinder's curvature (green), to the sphere's curvature (blue), and then back to the cylinder's curvature (green - though this part of the intersection would be blocked from view by the sphere). First figuring out the relevant curves helps give us the pieces to this intersection, but we still need to figure out the specific way in which they're arranged. For that, we have to rely on our spatial reasoning skills, as we're certainly not going to break out all of the complex mathematics to figure this one out.

Anyway, there's much more to this lesson submission than the form intersections, so let's move forward. Your work on these object constructions definitely develops over the course of the set. Earlier on, you definitely dabbled with the idea of working with greater precision, subdividing your initial bounding box to find your major landmarks and such, but I'dsay that it isn't until your cellphone where you really start fully grasping just how far those subdivisions can be pushed.

The concept of precision can be a somewhat confusing one, easily mixed up with accuracy, but they're fundamentally different concepts. Accuracy is about looking at how far off you were from making the mark you intended to make. Precision, on the other hand, comes from having defined your intention first. When we use things like subdivision to establish the specific location for, say, a button or a switch, we're making our intentions clear. This element should be positioned 2/3 along the length of the object, or that element should butt up against the halfway point and extend for another quarter of the object's total length. We plot out these specific footprints first, before actually building out the element in question. We may miss a little here and there, slipping up and having a line fall in the wrong place, but our approach is still precise, if not accurate.

The more we subdivide, the more we pour our time into those kinds of procedures, the more precise our drawings become. It also inherently lends itself to increasing accuracy too (just as the ghosting method as a whole requires us to put time into the planning phase, which in turn reinforces our accuracy by just ensuring that we know what we're aiming for rather than making half-random marks), but that's just a happy side-effect.

From that cellphone onwards, you definitely show a greater propensity for investing more time into each construction. If we skip back to the upright piano, a good example of skipping a step that might have provided more precision is the fact that you positioned the foot pedals around the center line, but didn't establish how far out to either side they should sit. You eyeballed that instead, when a better approach would have been to at least put a pedal down on one side, then reflect it across the center line as explained here to keep things centered. Is there anything visually wrong with your piano pedals? Not really. To the naked eye, they look just fine - so one might argue that they're accurate, but not precise.

Another thing that seems to diminish as you push through the set is the use of somewhat arbitrary black shapes. We could see these in your earlier drawings (like the chair and TV), and it continues up to your clothes iron, but after that you seem to shift more towards focusing on using those filled black shapes as cast shadows. This is entirely correct - in this course, you really should only be using filled areas of solid black for cast shadows, because that's what the viewer's going to expect (given our limited use of tools here, and the fairly strict limitations to working in solid black and solid white). Don't worry about capturing any form shading (as explained here, it shouldn't be incorporated into your drawings for this course), and leave any sort of local or surface colour alone, treating each object like it's covered in the same flat white. Reserving those solid blacks for cast shadows will really help push the use of those cast shadows to help define the relationships between different forms in 3D space.

To that same point, avoid using hatching for much the same reason. It tends to read more as a sort of "hedge" (where we don't really want to commit to one side or the other), and will usually end up being used where the student wants to capture some form shading, but knows they shouldn't. We can see this in your webcam drawing, for instance, where it's clear that you know sticking to cast shadows is the right call, but you're still not 100% there.

Anyway! As a whole I think you're doing well, and rather than needing to redo things after your break, it looks like you've continued to improve even after coming back. I'll go ahead and mark this lesson as complete. Ultimately the focus and greater attention to precision you've demonstrated here will serve you very well once you reach Lesson 7 (which you're quite close to) - just remember that precision requires time investment, and when you get to drawing those vehicles - the cars especially - there will be a pretty hefty demand for your time. Some students panic and rush in response, but as long as you keep your head about you and focus on giving each construction as much time (and as much precision) as they individually require, you will do fine.