Starting with your form intersections, your work here is generally coming along quite well. I did catch a couple of small things that I wanted to correct, which I've marked out here, but as a whole you're demonstrating a very well developed understanding of the spatial relationships between these forms. The main thing to keep in mind is that when you're intersecting a cylinder/cone/other curved form with a box or other flat surface, you have a tendency to follow the alignment of the cylinder when picking the cross-section to use in your intersection line, but instead, it's the plane of the box that determines what cross-section to use. In both the cases I highlighted, the cross-section would have been set at an angle to the form itself, not aligned to the form's own minor axis. This diagram may also help.

Continuing on, your cylinders in boxes are still looking good, so I won't dwell on those.

Some students have a tendency to overestimate just what goes into the initial 'form intersection vehicles', but I'm pleased to see that here you've held to the core premise of the exercise. That is, to build the structures out with simple forms, arranging them as though they were a vehicle, but otherwise sticking to the form intersections exercise (in just drawing primitives). And in that regard, you've handled them quite well, with a lot of focus on the individual pieces being built up.

Of course, this exercise kind of takes us back into how we would work in a "reactive" manner from Lessons 3-5. Basically, if something ended up being too big, we'd simply move over where we intended to draw the next form, "reacting" to how the structure was being built out, basically working from the inside out. Then in Lesson 6, we introduced the concept of working "outside in" - defining the boundaries and measurements of things first, and then breaking apart the structure into smaller pieces contained within that volume.

The thing is however that this "outside in" approach does not involve starting with a big volume and then drawing the entirety of our object within that space. While we're working within the boundaries defined by our subdivisions, we're still building the objects up with smaller primitives. So for example, here on your coast guard boat demo, we can see how we're breaking the structure up into four major components. There's the base, made up of the red box and the blue triangular prism. Then on top of that, there's the orange box and the green box. We are constructing each of these one at a time.

I think you've done an especially good job of this with this train construction, where I can clearly see you blocking it all in with smaller primitives, all with flat surfaces, and then rounding them out towards the end (although as I've noted here the curve in blue is better than the one in red, as the one in red does not make direct contact to the scaffolding that supports it).

Generally most of your constructions are handled quite well, but in a manner that is not that uncommon with students, once you hit the cars, you end up relying less on what you know. I can tell that you end up panicking somewhat, because they're vastly more complex, and this impacts the manner in which you tackle the constructions.

The main issue is that you stop building up structures inside of your bounding box. Rather, because it's so complicated, you end up focusing a lot more on working from bounding box to subdivisions to "final drawing", without paying as much attention to the individual blocks and pieces you're using to underpin that construction as it gets built up.

Firstly, you end up forgetting this important instruction from the lesson. Now, I know that you were doing this throughout the lesson, but it really starts to become a problem here, because when we switch pens up and go back over our lines, we end up thinking of the drawing as being an underlying construction, and a clean-up pass. This is a common, and entirely valid approach in drawing in general, but it is not one that should be used here because it erodes the idea that we're building something up, bit by bit. Not redrawing or replacing, but always building upon the existing structure.

In your cars, your curves end up separating a great deal more from the flat/straight structures that are meant to support them. Where in that previous diagram I showed the greater separation between the blue and the red, here it goes much, much farther, and the relationships between your phases of construction become weaker. You estimate those "final" lines, instead of considering how they're built onto a 3D structure.

Looking at the orthographic proportional studies that you provided over discord, I can see that you did think about those flat planes/edges instead of jumping straight into curves, so you started out on the right track. One thing that I believe will help you in carrying that precision into your construction however is to actually try drawing these orthographic studies yourself. This is not something that is always going to be required, but I think it would help given the areas where you end up not investing as much attention.

One of the things about the orthographic study is that it's not just about making observations - it's also an important area where we make decisions. For example, if you take a look here, I've marked out a red line that marks the height of the window section's bottom towards the back, and a blue line that marks the same for the front. These are at different heights because the windows kind of slope towards the front of the car.

The green line I've marked in there is really the closest proportional landmark (1/4 down from the top of the car) that you've got, and so the window slant needs to be estimated based on those... which is tricky to do at best. What we can do here is force the red marker to where the green line is, and then find some other smaller subdivision to give us a close-enough height for the blue marker.

These aren't accurate (because we're fudging the numbers to bring them closer to "reasonable" subdivisions (we don't want to be working in 1/90ths), but they are precise because we are making the decision ahead of time. That's something that we can forget about when we're just marking on top of the photos - if we're actually drawing these proportional studies ourselves, we can make those decisions and generally be more aware of all the different elements that need to be decided upon.

That about covers it. As a whole, you're doing pretty well, but when you hit the cars (which again are understandably way more complicated than anything else), you end up panicking and forget some of the important points that I know you understand. You also may be underestimating just how much time these should take. If you're under the presumption that these should be completed in one sitting, you may want to reevaluate that.