Hi! Okay, let’s take these one at a time~

1. You can read about this here, or, if you prefer hearing about it, it's at 4:49. Basically, our brains don’t like to rotate things. If all we have is a single box in the center, we’re likely to go for a lesser range of rotation than we intend. However, if we define the range of rotation beforehand, by having the center box, as well as the further off box (that’s fully rotated), our brains will try to fill in the blanks, and we’ll end up with a full range of rotation. That’s the idea, anyway.

2. Technically, yes, but students aren’t expected to have that fine of a control at this stage, so the recommendation is to keep it consistent, and shallow.

3. Yes. The boxes are entirely unrelated, and the recommendation is to experiment with different degrees of foreshortening, with an emphasis on shallow foreshortening.

4. The first method is correct, provided you’re imagining the vanishing point, and not in any way marking it out on the page, or outside of the page. The second method is fine, for the moment, but not ideal. Basically, if the box is anything but a cube, one side of it may need to converge a little faster than the other, and it’s a lot harder to calculate this using that method. It’s fine to start off, however.

(Those points at the end are correct, and, by the way, are to keep in mind for the future. You’re clear to move on to the box challenge!)