Starting with the first section - the cylinders around central minor axes, you've clearly improved a great deal in getting your ellipses to align to those minor axes, and as a whole your ellipses themselves are coming out smooth and evenly shaped, which is great to see. There is one big shortcoming however that you don't appear to be aware of: you are only applying one of two major aspects of foreshortening to your cylinders, and as a result most of them look a little off, though the viewer may not be able to pinpoint why.

Foreshortening has two major elements to it, two different ways in which the closer and farther end of the form "shift". One is the shift in degree, which you appear to understand reasonably well (the far end gets wider than the closer end, as explained here).

The one you're missing however, is the more basic type of shift - where the far end gets smaller than the closer end. Across all your cylinders, you're effectively keeping both ends roughly the same overall scale, and in some cases the farther end actually (accidentally) comes out larger (like 149, for example).

The thing about these two kinds of shift (the degree shift, and the scale shift) is that they must be applied in roughly the same amount. You can't have no shift in scale but some shift in degree, as you've done throughout all of your cylinders here. Reason being, these shifts help communicate to the viewer the relative length of the form itself. If the shift is more significant, it'll suggest a greater distance between the ends, meaning a longer cylinder. If they're less significant, then it's going to be a shorter distance, and therefore a shorter length. If however one shift suggests a longer cylinder and the other suggests a shorter cylinder, you end up with a contradiction.

Whenever going through an exercise where you're to draw many of a single kind of form, it's a good idea to push yourself to explore many different variations - in orientation, in foreshortening, in scale, in dimensions, etc. - so as to push your understanding of that form as far as you can. Additionally, when working in 3D space, do not attempt to eliminate all foreshortening. Sometimes people purposely do this, keeping their sets of parallel lines (parallel in 3D space that is) parallel in the 2D space of the page, getting rid of the added challenge of keeping convergences towards vanishing points in mind by just doing away with the VPs altogether. Even if the foreshortening is very slight, you should still always be aware of the vanishing points/convergences of your sets of parallel lines.

Moving onto your cylinders in boxes, I'm seeing some further signs that you didn't really understand what was being asked of you - specifically because you're missing some of the line extensions that you were told to draw as part of the exercises. From what I can see, you only drew the line extensions for the box itself, not the additional lines defined by the ellipses (the minor axis line, and the two lines created from the 4 "contact" points). Furthermore, when drawing the boxes' own extension lines, you somewhat regularly extended at least one of the sets of lines in the wrong direction, extending them towards the viewer rather than away. Extending the lines towards the viewer is not useful to us - we extend the lines specifically to study how they behave as they move away from the viewer.

Now, the reason the additional lines from the ellipses are so important is because this exercise is largely about helping you develop a stronger intuition in regards to constructing boxes that are proportionally square on two opposite ends (since this is the kind of box that'll enclose a cylinder, and this ends up being quite useful in lesson 6). In the box challenge, we use the line extensions to study how our lines are oriented, and how they converge - we find where those convergences aren't entirely consistent, and over many pages of these boxes, we make alterations to our approach to improve that consistency, ultimately improving our ability to construct these boxes, and to judge those convergences by eye.

By adding the cylinder, we provide ourselves with three additional lines per ellipse. These lines, as explained in the instructions for this exercise, will only align with the box's vanishing points when the ellipse in question represents a perfect circle in 3D space, which sits along the surface of the given box. If the lines are off a little, it suggests that the ellipse isn't quite correct, and so we can make slight alterations to keep trying to bring it more in line, in turn improving our instincts in regards to drawing the boxes with the correct proportions.

Because you didn't use any of these additional line extensions as instructed, you did not develop in this manner. As such, I am going to ask you to do this second section - the 100 cylinders in boxes - again. Before you do, please take the time to go through the instructions again, and pay closer attention to what is being asked of you.