Overall your work appears to be largely well done, so before I get to what I hope will be a fairly brief critique, I'll jump right into your questions.

To your first question, unfortunately you're not giving me a whole lot to work with here (specific questions are easier to answer than general "I don't understand") but I'll try to explain the concept as best I can. The first point that I can clarify for you is that it's not a technique that is a good choice for all objects - rather, there are many objects (like your jar, which is symmetrical in multiple dimensions) that lend themselves to be defined primarily as simple boxes, and whittled down from there through simple subdivision.

Your iron, however, was a good choice for this technique - as would be something especially complex, like a shoe. Because of their complexity, it can be difficult to wrap our heads around how we might start with a box, and then steadily cut pieces away from it to reveal our desired result. Instead, in this technique, we use "cross-sections" of our object to help plan it out. A cross-section is a very thin slice of an object. We can make these cuts in any dimension, and they result in what is effectively a two dimensional shape - like something cut out from a piece of paper.

For especially complicated objects, we can take many cross-sections. In the computer mouse demo, we work with two distinct cross-sections. One, in blue, that represents the "footprint" - so if you took your object and traced all the way around it onto the page. Then there's the one in orange, that is as though we took a very narrow slice from the very middle of the mouse. Each slice allows us to focus only on two dimensions at a time, instead of worrying about all three dimensions at once.

We could also have chosen to add additional cross-sections here, to further simplify the problem of constructing this object. For example, while we have that orange middle-slice, we could also take slices from the left and right sides of the object (in red here I've marked in what a side cross-section would look like). With the mouse being symmetrical, these slices would be identical, and they'd help us further establish the structure, without ever having worried about more than two dimensions at a time.

The more of these cross-sections we build up, the more structure we have, simplifying the rest of the construction beyond that point, when we do start working in three dimensions at once. This technique also allows us to more directly apply our orthographic studies of our objects (where we can focus most of all on using subdivision to find where specifically each little element or feature of the object sits in proportional relation to the overall object (like, if a particular corner is 1/3 down the length of the object, or 1/2 way down, etc).

As to your other question, if you're trying to figure out how to position a completely arbitrary line, there aren't a lot of tools to help do this with precision - we're essentially stuck trying to, through analysis of our object, identify where each individual point/vertex of that line sits. Basically, like at the end of my last paragraph, we try to identify if a particular point is going to be halfway down the length, a third, a fifth, etc. Doing this for each corner/point along our arbitrary line will help us figure out what kind of subdivision to use in order to place those points in 3D space. From there, it's a matter of connecting the dots. This is also why it's best to always start a curving line as a series of straight lines first, as explained here. While a curving line won't have these kinds of specific vertices along its length to position with more precision, when we first represent it as a series of straight lines, it becomes easier to plan out.

As an example of this sort of "connect the dots" approach, let's take a look at the top of your spray bottle construction, where I've pointed out 4 vertices that help define the back edge of the structure. A, C, and D are all positioned quite specifically based on subdivisions of the original bounding box, but B is positioned a little more arbitrarily, resulting in a loss of precision. For this, we have two options - you can either look at your reference and, in an orthographic study, subdivide as much as you need to find the specific proportion (1/2, 1/3, 1/5, 2/5, etc.) that matches up with that corner, or you can... well... cheat. That is, lift B up a little so it falls right at the subdivision above it.

That isn't necessarily always going to be allowed, but for the purposes of this course, we aren't really that caught up in replicating our reference images perfectly. Our reference images are just sources of 3D information, which we use to construct real, believable, tangible objects in our constructions. In doing this, we train our brain to think in 3D space - this doesn't require us to perfectly replicate some photograph or real-world object, and so for the purposes of what we're doing here, this sort of shortcut is allowed - as long as it leads you to working with precision (that is, figuring out exactly where each corner goes and planning it out through subdivision ahead of time).

Now, when it comes to your actual work, you really have done a great job. While there's one or two odd cases where you didn't plot things out entirely (like point B in my previous example), they are few and far between. Overall, you've worked with considerable precision, and have demonstrated an immense degree of patience and care as you analyzed and reconstructed each of these objects on the page. Each one feels solid and believable.

Similarly, your form intersections demonstrate a really solid understanding of how these forms sit in, and relate to one another in 3D space - specifically as demonstrated through your expert use of the intersection lines. You're showing a lot of comfort both with the intersections between the somewhat easier flat-to-flat intersections (like boxes intersecting with other boxes), as well as the much trickier curve-to-curve intersections (like spheres intersecting with cylinders). Just remember to take a little more time when drawing your hatching lines, so they all stretch from edge to edge across the plane, rather than stopping short.

You've done a great job, so I'm going to go ahead and mark this lesson as complete. Keep up the fantastic work.