INANE (I Am Not A NASA Engineer), but perhaps 4 or better yet more legs arranged in a tetrahedron or analogous polyhedron would better ensure the lander did not get stuck inside a crevasse; an inner rotating body could then bring the lander into a working position. The more the legs, the higher the chance the lander could 'roll' to a lower ground when initially bouncing off the surface, but the optimal number of legs could be determined in advance by simulating the terrain (you wouldn't want it to roll all the way down a slope, ending at the bottom of a crevasse etc.).
Harpoons protruding from each leg could be fired automatically as soon as a stable enough position was detected, or manually when deemed safe; having more legs available could only improve chances of getting a safe fixation.
Philae had a harpoon to anchor itself to the surface, but ended up malfunctioning. I assume had it actually fired, the engineers balanced the need for extra landing gear and scientific instruments quite well.
Harpoons protruding from each leg could be fired automatically as soon as a stable enough position was detected, or manually when deemed safe; having more legs available could only improve chances of getting a safe fixation.