When I talk about the twist material being opaque being a problem, I’m not saying it blocks all the light. I’m talking about how the light comes off in a sphere from all points throughout the energized gas in the bulb. It doesn’t just go straight out laterally, but in all directions. That’s why you would be able see the whole bulb lit and not just the part perpendicular to the surface of the bulb. So the light rays going nearly straight out perpendicular to the surface of the bulb will have their full effect on that water perpendicular to it, hitting water until they eventually hit the sides of the tube, just like any UV sterilizer.
But the light photons not going out nearly perpendicular (instead going out at every angle), which is most of the light, will have a shortened path before they hit the twist material, not reaching their full effect of hitting as much water as possible before hitting the outer wall. This will waste a significant fraction of the total UV produced by the bulb. Think of it as the twist material partially shading the light that would have otherwise been coming from above and below the level to hit the water at a certain point.
As far as the concern about not all parts of the water getting a consistent flow and dwell time in an open design, that is true. But the tubes are designed so that on average water flowing through it will get the appropriate amount of radiation. Not perfect but a reasonable estimate. A lot better than blocking a lot of the light.