I'ts not just similar to the polytwister gears idea, it's actually the same thing.
The decomposition into two orthogonal circles is just the Hopf fibration of the digon, which is just the simplest case of a polytwister gear.
The interesting thing about this decomposition is that rotation in one ring is equal to twist in the other ring. That is, if you attach teeth in two orthogonal great circles around the glome, and lock gears into either ring, then rotating one gear causes the corresponding ring to spin in its plane, and that causes the orthogonal ring to twist in-place (i.e., rotate around the great circle) -- the teeth don't move but they will spin. Similarly, if you make the teeth ridged such that you can twist them in place, then doing so will induce a rotation in the orthogonal ring.
Making the teeth smooth (i.e. they can freely spin in-place) will allow you to use the glome as two independent gears simultaneously. In this case, one ring can rotate independently of the other; the other can stay put, but the teeth will experience wear as they spin in place.
But my thinking is that the rotation-twist correspondence can probably be exploited in other ways, such as making a gearbox, or something similar.