It seems like that would massively increase complexity - right now the energy is collected at one point, having it collected all around the tower means more leakage and way less maximum energy.

I think you need to use the entire distance from the sun, which means the few meters between the mirrors is a miniscule difference

60 meter (200 ft).

https://en.wikipedia.org/wiki/J%C3%BClich_Solar_Tower

I don't think this is what they are directly aiming for with this specific one, but this general idea of a large field of reflectors and a central receiver can be built to store heat for a while. That lets you use solar power overnight

This sort of solar power generally doesn't compete successfully with regular solar panels in cost, and obviously it can't just be set up on a bunch of random roofs like panels can. Still, it's potentially a very useful technology if someone does work out how to make it a bit cheaper or more efficient, and it still generates clean power in the process too

Photovoltaic solar panels turn light directly into electricity. This thing bounces light to a place where it can be turned into heat. The heat boils water, and the steam turns a turbine. The spinning motion is converted into electricity.

Steam power is a lot older than photovoltaic cells. Nearly every other form of power generation is basically just spinning a turbine. Nuclear, coal, and geothermal heat up steam. Hydro and wind get nature to spin the turbine. With steam power, all your problems with efficiency are mechanical. How do you reduce friction in the bearings? How do you get more power out of the steam? Whereas developing more efficient PV cells requires a deep understanding of subatomic physics. On the other hand, it's difficult to think your way out of physical limits on steam power, like where does the waste heat go? The future of PV probably still has surprises left for us.