The idea is to construct a vehicle out of 2 or more modules, connect them with cables and spin the entire assembly. The primary goal is to provide artificial gravity with a relatively low spin rate.
Another advantage is to allow a nuclear power source with a minimum of shielding. Specifically, radiation decreases by the square of the distance, so the mass of a long cable is far less than the amount of shielding materials needed to produce the same decrease in radiation. In fact, any shielding would be located around the crew module to double as protection for solar and cosmic radiation.
The distance needed to provide radiation protection determines the (minimum) cable length, and desired artificial gravity determines the spin rate.
Since interplanetary voyages will take a very long time, low thrust high mass-efficiency ion drives would be appropriate, with the nuclear reactor being used purely to provide electricity (in other words, a large RTG).
The main disadvantage of this design is the large angular momentum. To minimize angular momentum, one should place as much mass as possible at the center of rotation. This include engines, reaction mass, the lander vehicle, the main communications attenna, and most of the supplies.
While a system of 3 or more modules is subject to pendular oscillations, these can be parametrically damped using "shock absorbers" between the cables and the modules. Multiple angled cables would also help, but are not strictly needed.
Our design now consists of three modules: a crew module, a nuclear power plant, and a long hub. A pressurized elevator shuttles between the crew module and the hub, used primarily to transfer supplies. It can also double as an air lock.
It is not necessary to pressurize much of the hub. All that is needed is a small docking module to interface with the elevator, lander and a couple of storage compartments. Storage compartments can be arranged down the length of the hub and moved to the docking modual via rail arms (a robotic arm mounted on a track down the length of the hub).
Because of the spinning, it would be best to use a phased array for the main communications antenna instead of a parabolic dish. Also, the ion engines need to be gimbled to point in almost any direction. During the planetary orbital escape and insertion, the thrust direction will change faster than one can re-orient the spin. However, most of the time one can lock the engines along the spin axis, precessing small changes in spin orientation (without wasting reaction mass) by turning the (off-center) engines on and off in phase with the spin.
(An alternative design would be to mount engines pointed in all directions and use whatever engine is pointed in the right direction at the time.)
A logical hub design would be to put the lander and docking module on one end of the hub, the antenna on the other, and the engines on arms off to the side near the center of gravity, perpendicular to the cables. One can also use the engines to change the spin rate. The cables would be attached near the ends of the hub for stability.