The marble is not cooler than it's surroundings. It only feels cooler because it conducts heat away from it's surroundings more quickly than say a butcher block (wood) counter.
But this is the important point since reptiles are able to lose or gain heat quickly. It makes sense to me that a reptile should be able to thermoregulate by moving onto a slab of marble just like it can thermoregulate by moving into a water bowl, albeit without the risk of excessive soaking. Water is more effective since a reptile can typically submerge in it, which increases surface area exposed.
Your idea is still valid, just for different reasons that you are considering.
A scrap piece of slab (thick) marble would be much better than a tile, basically because the surface area relative to the mass would reduce radiation gain. And scrap marble can often be had for free or at least very cheap from stone yards.
And you might also think that stacking marble tiles two or three thick will have the same effect. This won't work because of the rough texture on one side of tile, which allows mortar etc. to stick. This would also reduce conductivity between tiles.
Lastly, there is some evidence suggesting that some species of reptiles do not instinctively recognize surface temperature differences and are not able to thermoregulate accordingly. It could be that some species will burrow or soak before crawling onto a slab of marble, even if it would work. Simply put, some reptile associate being out in the open during the day with "warming up" not matter what the actual circumstances are.
-----
Currently keeping:
5.8 Gonyosoma oxycephala (Javan, mixed colors)
1.1 Philodryas baroni
1.0 Rhodesian Ridgeback