If you're trying to sight down the centerline of an alignment grid, having two images makes it possible to get precisely lined up. The printed line and its reflection are either directly on top of each other or they're not. This degree of precision is impossible without dual images.
I spent some time with my TTB protractor today and I see what you mean. If the lines were thinner and they got rid of those concentric squares, I think it would be a better tool, though.
If using a universal protractor, absolute precision in pivot-to-spindle distance is unnecessary. That's the whole point of an aimable protractor and headshells with slots. Pivot-to-spindle is a hair off? No problem. The protractor will automatically produce a pivot-to-stylus dimension to compensate. If the stylus is square at both null points your triangle is correct, by definition.
ok...I get it. Finally. It only takes 6 or so times to sink in.
BTW, if you're using a Baerwald protractor with an OL arm, you may have better luck mounting the arm at ~220mm than the stock distance of 222.76. Regas weren't designed for Baerwald and some cartridges aren't long enough to reach the Baerwald null points with the arm mounted at the stock distance.
I know we've discussed this before, but I've never experienced this as an issue (Shelter 501 mk 11 and a Denon 103R).