An interesting fact about this is that the moon's orbit is always concave to the sun -- I think that also implies it is never retrograde either, but I've not done that calculation.
You mean the Moon's orbit is always convex, and never concave. That's true. And yes, it implies that it never moves retrograde as seen from the Sun (but the opposite implication is not true).
Well, just for the sake of argument, and to give a starting point, let's stipulate that it's one Earth-sized planet, and two moon-sized moons, with the current size, shape, density, etc. of the current earth-sized planet and moon-sized moon. Would celestial mechanics allow those two moons to orbit around the planet?
I think the question is similar, although not identical, to this question: can three stars of comparable mass form stable systems?
Well, there are systems like that that are reasonably stable. The general rule of thumb for stability in such cases is hierarchy: if masses are too similar to establish a clear hierarchy, distances can do - meaning that two of the objects will be relatively close and the third one farther away.
In the case of one Earth and two Moon-sized moons, there are two main ways to arrange that: either one of the moons could orbit much lower than the other, or the two moons could orbit each other closely at a high orbit around the Earth. (This is of course only a basic recipe for long term stability; a general guideline of where to start further search for actual self-correcting orbital resonances.)
The presence of the Sun adds another layer of complexity, but its great mass and distance already offers hierarchical separation; in essence, it makes the situation conceptually equivalent to a quadruple star system (with a certain ratio of constituent masses) - and even those are in general known to exist.
