Starting with your cylinders around arbitrary minor axes, for the most part you've done a good job here, with one important mistake.

You're drawing your ellipses with confidence, and your marks come out evenly shaped and smooth, which shows a good application of the principles from the ghosting method. You're also quite fastidious in checking the alignment of your ellipses, picking up fairly small mistakes at times, rather than complacently drawing along your intended minor axis when it's close enough.

As for the important mistake, there's two parts to it. Firstly, the assignment did ask students to vary their foreshortening quite a bit here. While there's some variation, there's not a whole lot. Furthermore, in some cases like this one from this page, you actually appear to be eliminating the convergence of those lines altogether, drawing them as though their vanishing point is at infinity (in the manner explained back in Lesson 1). This however is not something we can arbitrarily assert. We only control the intended orientation of the form, and the vanishing point(s) follow that. If the set of edges in question runs perpendicular to the viewer's angle of sight, without slanting towards or away from them through the depth of the scene, then the vanishing point ends up at infinity. In any other orientation however, we end up with a concrete vanishing point.

Given that this challenge has us rotating our cylinders freely and randomly, the chances of such a perfect alignment is pretty slim, to the point that we can assume that it's not really going to occur. That said, having one or two instances of this in a set is not a big deal. You have a bit more than that, but more often you have cylinders that converge very gradually. That's fine, but I would have loved to have seen more dramatic foreshortening throughout.

The reason I asked for more variation is to catch a specific kind of mistake - specifically one where the two shifts that our ellipses undergo (the shift in overall scale due to the convergence of the side edges, resulting in the far end being smaller than the closer end, and the shift in degree where the far end is wider than the closer end) act independently of one another. There are a few places I can see this - for example, this one on this page where the farther end actually gets narrower instead of wider, this one on the same page where it maintains roughly the same degree while getting noticeably smaller, this one from this page and this one from this page which are much the same. There are of course others as well.

So, keep in mind - both of these manifestations of foreshortening must occur in tandem, because if they don't, you'll end up with a contradiction which the viewer will notice, even if they can't place it specifically. Both of these 'shifts' serve as the visual cue we use to identify whether the length of the cylinder on the page is about how long the form actually is, or whether a larger portion of its length exists in the "unseen" dimension of depth.

Continuing onto your cylinders in boxes, by and large you've done quite well here. You've drawn your boxes with ample variety in foreshortening, and you've applied those line extension techniques effectively, which is at the core of the exercise. This exercise is really all about helping develop students' understanding of how to construct boxes which feature two opposite faces which are proportionally square, regardless of how the form is oriented in space. We do this not by memorizing every possible configuration, but rather by continuing to develop your subconscious understanding of space through repetition, and through analysis (by way of the line extensions).

Where the box challenge's line extensions helped to develop a stronger sense of how to achieve more consistent convergences in our lines, here we add three more lines for each ellipse: the minor axis, and the two contact point lines. In checking how far off these are from converging towards the box's own vanishing points, we can see how far off we were from having the ellipse represent a circle in 3D space, and in turn how far off we were from having the plane that encloses it from representing a square.

In applying these line extensions and analyzing them as fastidiously as you have, you've clearly shown a good deal of improvement over the set, and while there is certainly plenty more room for refinement, what you've achieved here will serve you well into Lesson 6 and beyond. So! I'll go ahead and mark this challenge as complete.