Their accuracy is fairly impressive. The first point you made explains the discrepancy all by itself. The other two should not be considered.
For 2) : It does not matter if a cubit is standard, as long as the same forearm is used to measure the pool and the rope.
For 3) : It is trivial to mark a circle using a piece of rope and a stake. The circle would by no means be perfect, but neither would it be a 'crude approximation' - it would be accurate enough to get a more accurate measurement of pi. Assuming a non-stretchy rope and accurate measurement of rope lengths, they could have gotten a few decimals right.
Of course, there's the issue with them probably not being very good at dividing a cubit into sections. If their rope ended up being 31 cubits and a little left over, they might just say it's 31 cubits.