I am not NSA, but I think their idea is something along the “once Shor’s algorithm is practical for key sizes of hundreds of bits, it’s only a matter of a few years of engineering effort to make it feasible for thousands bits long keys”. In other words, there is no sufficient safety margin for larger keys.