Couldn't agree more with your vision of a rational strategy for robotic exploration of Mars, which parallels what I've been thinking for a good while.

Pathfinder worked, and the airbags especially proved their value. Unlike the Lunar Module, there's no pilot aboard a Mars lander to steer it around obstacles. It's simply prudent engineering to design the landing system on the assumption that the first touchdown point will be occupied by a crater or boulder. The airbags let the lander "seek" a proper landing spot.

(One unique benefit of the airbags is they would encourage landers to end up in a crater, like a golf ball in the cup; and we'd get the chance to look at the exposed subsurface, accumulated dust, and possibly signs of fluid collecting in the depression. Instead of being feared, craters would be prized.)

Every field of engineering, outside of aerospace, apparently, recognizes the value of standardization and mass production. NASA could let a contract for a hundred basic lander chassis, designed with attachment points to plug in the instruments to fly on a particular mission. With a standard interface, private labs and industry can design their own instruments. The communication, navigation, and propulsion systems don't vary from mission to mission. By using larger boosters, we could lob a half dozen landers at Mars with each launch, carried on a "bus" to Mars and then released once in orbit. The bus could then act as a comm relay and add to the constellation of orbiters watching the planet.

NASA loves R&D a little too much. I think we know enough about the engineering problems to just get on with it. Science is suffering while we indulge the engineers.

--Robert