Skip the gravel, 3/4" minus crush about 6" worth is the way to go
If its real bad dig it out and put down a layer of "road base" first (basically a coarser crush)...
Yep, that's along the lines of what I was thinking, especially on the "coarser crush" part. Ryan, you obviously have done some this stuff before.
I also agree with your prediction that this layer of gravel will last longer than the last one.
Brent, another part of the equation is getting the material uniformly compacted. (And like Ryan said, crown the road, keeping the center higher than the edges.) Until it's compacted, it'll continue to move around. Although the wheels from vehicle traffic will compact it somewhat, it's really only getting consolidated in the tracks...the areas on either side of the tracks will remain relatively soft. (That's
relatively soft, compared to the packed parts getting all the vehicle traffic. It may still feel or look hard.) Anyway, when more traffic comes along, it smushes the material in the tracks to the sides, and eventually you end up with ruts again. When you compact the areas beside the tracks, those sides greatly help support the track areas themselves. A vibratory steel-drum roller would be ideal, but it's unlikely you'd want to pay the cost to rent one. Another workable alternative would be to run over it repeatedly and evenly with a rubber-tired vehicle. Start on the edges and work your way to the center. And the higher the air pressure in the tires, the better results you'll get. (It has very little to do with the weight of the vehicle; the air pressure in the tires is the maximum PSI of downward force you'll be able to apply to the ground.) I'm betting Bob has the highest-PSI tires on your compound.
Also, it's important for the material to be at the proper moisture content when you're compacting it. The more fine particles in the material, the more critical that becomes. The trick is, knowing when the material is at that optimum moisture requires either experience or lab testing. Every variety of dirt or gravel will have its own optimum moisture. To illustrate this, think of making mudballs out of farm dirt. If the dirt's real dry, you can't form it into a ball. Add a bit of water, and all of a sudden you can form the mud into a ball, but it crumbles. A bit more water, the the crumbling stops and you can actually create a throwable ball. Add even more water, and before long it's too wet to form anything but gloppy mud pies. Somewhere in the middle there (when you can pack a good firm, mudball) is the optimum moisture. Sand and gravel are a bit different, since you'll probably never be able to form it into a ball, but having some moisture in the mix when you compact it is still important. (The water serves as a lubricant to help the particles squeeze in next to each other and lock in place.) In general, you want all the surfaces of each particle to be wet, but no appreciable amount of free water.
Sorry for the long-winded explanation...it's one of the side effects of having a soils engineer for a dad and having worked in the materials testing and earthwork business for quite a few years.