I'm not sure if this view undersells aero or rocket engineers more.

Rocket engineering uses an immense amount of both modelling and physical testing. No-one says "Well I've got the Tsiolkovsky rocket equation, so let's go to the moon!"

I'm not even sure what the bar of 'understanding' is here - the fact that we have iterated and improved on powered flight as much as we have necessarily means that we understand it on a deep level, let alone the fact that we can create excellent models that predict what will happen to a wing in different situations.

At what point would you say we do understand flight?