I used to work with aerospace composites. Glue turns out to be a very good way to stick things together (I know, shocker). The trouble with screws in this application is that they are basically point forces, and they work by holding the action to the stock with friction, which is notoriously unreliable. We were forbidden to rely on friction to take shear at work - we had to have a pin, lug, or other positive mechanism to take shear load, and bolts don't count for other reasons. For that to work consistently, you need a very tight tolerance fit and the screws need to be tight. Over time, gunsmiths have figured this out, and that's why we have pillars and bedding jobs. The recoil lug is as much a way to locate the action as it is a constraint on recoil. This works, of course. But it's prone to degradation if disassembled frequently. Solvents, oils, etc can get in there.
A proper glue in with the right adhesive and bond thickness (the optimal thickness depends on the adhesive - we used to use wires to set it precisely) is incredibly strong in shear - I would guess it's stronger than the stock. A recoil lug in that case is totally unnecessary. If you read Harold Vaughn's book, at least some of recoil force in a properly bedded rifle will go through the recoil lug. This unbalanced force, in turn, drives barrel vibration in the vertical plane. A glue-in reacts the recoil force at the bond line, which is closer to the bore axis, and will impart a smaller moment to drive vibration.
I'm not willing to say that's automatically better, but it is different. A rifle with no driving moment ought to be very consistent (this is an ideal that does not exist). But a little may be desirable to exploit positive compensation. Quantifying that is difficult.