The main thing the box challenge has you work on is angling 4 lines such that they converge towards the same point. And so, while it's certainly not glamorous or exciting, you improve by thinking about how each line needs to be oriented against the others of its set, as you draw them. And you do that, a lot. Tediously thinking about all the lines of the set simultaneously, so that you can gradually shrink your deviations and improve the consistency of the convergences.

So yeah- your understanding of the goal is correct, except there's no real expectation that your boxes are going to be perfect after 250. Two hundred and fifty is an entirely arbitrary number. What matters is progress, and what forcing yourself to think about these convergences as you rotate your boxes arbitrarily in space does to develop your brain's internal model of 3D space.

I can tell you now - I can't draw a perfect box freehand either. And that's not a problem, nor was it ever the goal. But I am capable of constructing complex objects that fall within an acceptable margin of error, by combining different primitives and considering how they relate to one another in 3D space. This is the first step towards that.