Depends on what you mean by current approach as many different methods are in use.

Single junction solar cells are limited to 33.16%. https://en.wikipedia.org/wiki/Shockley–Queisser_limit

Increase the number of junctions and that goes up. Ultimately with infinite layers you can’t beat ~68.7% on earth and 86.8% when much closer to the sun.

These efficiency limits are true for P-N junction solar cells. There are other solar cell types which can ultimately achieve higher efficiencies, but they are currently lab curiosities with low efficiencies.

That’s the kind of thing I was trying to get at with my “current approach” comment.

Maybe solar cells based on chemical reactions or [insert science things here] can do far better, but right now semiconductors is what everyone sells.

What does closeness to the sun have to do with it?

If it's a question of the intensity of the sunlight, can't you just focus it with a lens?

Concentrator Photovoltaics (CPVs) exist but are expensive not just because of the lenses but because you need a tracking system to follow the movement of the sun and adjust the lens constantly to keep the light focused.