Hmm, well, that's not completely out of line --- my experience with Apple gear is that it will get a reasonable fraction of the theoretical wireless download speed, but upload is much slower. I've seen that with a few different wireless APs so I think it's Apple's fault not the APs'; they must be configuring their wireless cards with emphasis on battery life not transmit speed.
Having said that, I think you could likely do better than 562 down with a well-optimized setup. To start with, 4x XT8s is a whole lotta wifi, maybe too much; they could be interfering with each other. What sort of space are they spread across? If you turn off all but the main unit, and do your speed tests from say 10 ft from that unit, do the numbers improve? The rule of thumb is to use as few APs as you possibly can.
Wireless backhaul is also a performance gotcha in a lot of cases, which is another reason why I want to know what results you get when connected directly to the main node with no backhaul to be done.