I don't know if a single CEV design would be a good idea for orbital, lunar, and Mars missions. There is room for a great deal of commonality in some systems and software, and I completely support that, but a CEV required for a very short term mission to the moon just seems too small for a multi-month Mars mission.

I've heard that one corollary to Occam's Razor is to simplify as much as possible, but no further. If there was to be only one CEV design for orbital, lunar, and Mars missions, then it makes sense that the design be built to the most rigorous requirements - Mars. Still, going that route just doesn't make good engineering sense to me. As with clothing, "one size fits all" seldom works well. A CEV built for Mars could well be gross overkill for lunar and orbital missions, needlessly driving up costs. A CEV built primarily for lunar missions wouldn't be up to a Mars mission for a variety of reasons (life support, thermal issues, etc), and a CEV optimized for orbital missions is unsuitable for anything else.

An aviation analogy to the differing CEV requirements is what the military is trying to achieve with the F-35 Joint Strike Fighter. They're working to achieve high levels of commonality in propulsion, avionics, weapons, software, and similar areas. However, the Navy version will have a beefed up structure and larger wings to enable carrier operations, and the Marine/British version will have STOVL capabilities to meet those service needs.

In a similar vein, there can be a lot of commonality between CEV versions for orbital, lunar, and Mars missions in things like mission software, TT&C, basic life support (scaled for different mission durations), attitude control, navigation, and the like. Each mission would need its own thermal control system and perhaps electrical system design. Basic structural design would vary somewhat, but there is still room for a great deal of commonality between the different mission profiles, perhaps on the order of 60-75%. Trying to make one design be all things to all people is how we ended up with an overly expensive and overly heavy design like the Shuttle (and the F-111). Let's not make that mistake again.