Basically, the Retina model – where the LCD panel has one resolution, and the Displays {Preferences/Settings} has a different one – is a way to get around the issue that applications have almost always treated "pixels" as synonymous with "physical size." In the old days, when there were fairly small variations in DPI, there was not a lot of pressure to change this model. So you might see old 15" MacBook Pros offered with two different screen resolutions – with the understanding that if you chose the "hi-res" screen, you'd see more text at one time, but it would be smaller.
With displays that have Retina-level PPIs, the idea of using all of the pixels to cram more and more stuff onto the screen becomes untenable. That would be true even if there were no Retina "resolutions" in Display {Preferences / Settings} – even if there were more direct text/object size controls, instead.
There is an inherent tradeoff between
- The physical size of a screen,
- The physical size of individual text strings and objects ("stuff"), and
- The amount of "stuff" you can cram onto the screen ("workspace")
The physical size of individual text strings and objects matters because we don't all have microscopic vision like Kryptonians do in the comics. So with high-resolution screens, it makes sense to put some or all of the extra resolution towards more detail rather than towards workspace at the cost of readability.
Think of this way: for a long time, it's been possible to print books with much more PPI resolution than a Retina screen has. Yet somehow, publishers aren't rushing to put out books printed entirely using 3-point fonts – and people aren't rushing out to buy them.