Mcree just introduced something along the lines of an AR barrel extension for 700s in the new chassis. Takes any std AR handguard.
Manufacturing is really what has kept many of the ideas you ask for out of the market. It is a shame that the AR-company-that-shall-not-be-named (sounds like LC Offense) thread is gone, because while jackassery occurred, there was some interesting material there. The owner claimed to have spent how much bringing his product to market? $5k? 10k? $100k was what I remember, which really is pretty par for the course. If you want a quality item it is going to take time and money to develop. This cost is to be recouped by sales. Which along with facilities, material, labor, insurance, etc., etc., etc. constitutes the unit cost. Then add your profit. If you bond the action to the chassis then you decrease your modularity. High modularity means the customer can get more of what they want. This translates into higher sales as you increase the scope of your market. Ever hear of someone who said the AICS didn't fit them? While AI has used bonded actions for a long time, that does not mean that they are ideal for every user (finances not considered).
While the barrel extension idea is great, and is gaining traction, there really needs to be a single industry standard design, granted once the evolution of the concept has furthered. Otherwise, you run into issues with GS's that aren't familiar with it, make mistakes, sloppy tolerances, etc. The 700 action is simple. It has few parts. While not always true, a simpler design comes hand in hand with reliability, as there are fewer things to go wrong/break. For instance, to have a straight-pull bolt you're going to have to increase the number of moving parts, moving parts that all move relative to each other. Everything that moves experiences friction. Increase the friction coefficient via environmental conditions and most likely you have a design that will fail before the simpler design. Granted, this is all in theory, but these theories are not taught for naught.
As for factory-true actions, that is laughable. You've probably heard the phrase "it can be done well, fast, or cheap; pick two." If you've ever worked with machinery you understand tolerances. All machinery has their own. Increase the speed of a process, increase backlash, decrease precision. Decrease backlash by decreasing the mechanical tolerances of the machines parts, you're going to increase the chance of something breaking down/increase maintenance costs. Buy higher quality to avoid both, well you automatically increased your costs. While Remington's QC as of late is deplorable, to ask for custom action precision for the same price is asinine. They could just do it, and increase the costs of all their rifles, most of which would go unappreciated, and they'd be throwing money out the window. There's a reason why custom actions cost more. Why truing an action costs money. It is more expensive to decrease tolerances on parts, period.
Most of your points for evolution I agree with, as there is always room for improvement. That is true, and someone will have to be the guinea pig for it. We also have a vast knowledge of the existing systems, some of the peculiars are quite fine, and the gunsmiths who know them really know how to make a working rifle. And at some point in your argument you transitioned from advocating that we demand evolution as a community of consumers and started saying that you simply shouldn't have to pay as much. The latter is purely at the whim of market conditions and microeconomics.
Take a look at American Rifle Company's new offerings, they seem to be on the forefront of action evolution now.