I agree with your conclusions. Just a nit: the r² increase in surface of a sphere that makes this so much bigger an area to scatter the satellites across also a

I agree with your conclusions. Just a nit: the r² increase in surface of a sphere that makes this so much bigger an area to scatter the satellites across also applies to the field of view of telescopes and thus cancels out. Edit: Clarification. The above is only true if having a satellite in the field of view is enough to be a problem.