What A.T. is saying, I think, is that you don't need any periodic motion (or angular velocity) for the bulges to develop. As long as the body is in free fall in a central gravitational field, there will be two bulges. For what it's worth, any such motion will be technically speaking an orbit - even falling along a straight line on a collision course towards the barycentre is a degenerate orbit. Or being launched straight up to infinity, with escape velocity. But the reason to avoid thinking about orbits here is that it tends to muddle the water with intrusive intuitions, like assuming it has something to do with angular motion, centrifugal forces, periodicity etc. All you need is free fall and a gravitational gradient.
As they say, a picture with eighty-ish words is worth approx. 1.08 pictures. Or somesuch.
It doesn't matter how exactly the body is moving in the gravitational field, as long as it is free falling.
Now, if we literally nailed down the body (to the fabric of space-time? aether? firmament?) through its centre of mass, we'd still have an inertial frame as in #3, only with reaction force at F0. The proximal side would be under tension, the distal side under compression (i.e. one bulge only). Which is why letting oneself fall into a black hole leads to spaghettification, but trying to hover above the horizon leads to pancaking.
In yet another words, a slinky toy held by its centre of mass extends from one side only, but if we let it go (in a vacuum, with sufficient gradient to notice), it extends from both sides of its CoM.
If, however, we nailed down the body with a homogeneous gravitational field, as DrStupid suggested earlier, that magically works on the body only, and not on the source of tidal gravity, we would then be able to do the same subtraction as in #5. Since it would act with the same force on every point of the body. The gradient of the central field would then produce two bulges again.
Incidentally, the bulges in general needn't be aligned with the direction towards the centre of the field, but that is tangential to the topic at hand, and has to do with rotation under the bulges and the delayed response of the material to deformation. We can always consider a case with no relative rotation to simplify.