This is really facsinating to think that there might have been such large-scale uniformity in the early universe.
Most of the problem of comparing this result to theory is not that models have failed to predict it but that nobody (as far as I know) simulates this sort of stuff. I don't know of any large cosmological simulation which actually tracks black hole spin. It's only recently that cosmological hydrodynamical simulations have come along, even still accretion is below the resolution of these simulations so semi-analytic prescriptions are used. While there is clear prediction in simulations that we should see intrinsic alignments in galaxies (which is observed) simulations just don't include this level of detail.