Fun speculation: Maybe we started with no crossover (which gives avoidance behaviour, keeping the organism free-swimming). This still works for a while as the axial angle between eyes and muscle groups increases, so there's no real penalty for having a bit of a twist. As the twist increases, it starts acting a bit like a discriminator, where we avoid small things less than large things, which seems good if we want to eat small things. Past 90°, we start spiraling towards things instead of away from them, which admittedly makes us crash into large things more, but we can chase moving things. Hunting has evolved!

But there's a delightful span between seeing someone post something on HN unrelated to AI, Cryptocurrencies or startups selling VS-Code extensions and the moment when I satisfy myself something outlandish (like space starfish) hasn't happened. During that time, all things are plausible.

[Edit. Which is not to say I disparage or discourage posting cool things you've found on the net. That's kind of what many of us are here for.]

Where there is uncertainty is about what has happened to the head.

According to the supposition presented in the linked article, most of the head has preserved the same orientation as in most other animals, either because it never took part in the rotation of the rest of the body, or because later it has made an additional half-turn rotation, which has brought it back to the initial orientation. Among these 2 alternatives, I think that the first is more plausible, because in that case the environment would have retained a constant orientation with respect to the principal senses, even if the posterior body rotated.

While this theory supposes that most of the head of the vertebrates has the original orientation, also all competing theories must accept that at least a few parts of the head either have never participated to the body rotation, or a subsequent rotation has brought them back in the original position, e.g. the mouth.

It just made me think, whether it's related or not, shouldn't the "natural" resting position of human heads be "upwards" like the rest of the animal world?

I mean if you consider the worm/fish template, it's

    o--->

    o--->
     | |

    o—--v