Unless you were able to produce the fuel on the moon (which remains to be seen), it'd be a lot more efficient to use the lunar orbit rendevous concept. The rough numbers are that you have to reach about 4,000 MPH to achieve lunar orbit and 6,000 MPH for lunar escape velocity. In addition to all of the extra weight for propellant you'd need for that extra 2,000 MPH, you'd be carrying the heat shield, parachutes, and other mass from the moon's gravity well.

What I'd like to see is a vehicle designed as a sort of tug to travel between Earth orbit and lunar orbit. It would be designed to be fully reusable. From a propellant standpoint, the optimum solution would be to use some form of aerobraking to bleed off the extra velocity and enter Earth orbit. The extra velocity is significant, on the order of 7,500 MPH. Aerobraking might be able to get rid of most if not all of that. It would be maintained and refueled in Earth orbit, fitted with the next payload, and flown back to the moon at the next launch window.

I want a tug able to leave Earth orbit with a payload on the order of 25 metric tons, and more would be better. The payload would be deliverable to lunar orbit. For the first trip, it would be carrying the lunar lander, propellant, and perhaps a module for surface habitation. Subsequent trips would carry additional propellant and surface equipment. I don't see a big need for returning much from the moon but it wouldn't hurt to include some capacity for doing so.

The lunar lander would be designed to repeatedly fly from lunar orbit to the surface and back. I envision something similar in concept (if not appearance) to the Sikorski CH-54 Skycrane helicopter. It would be designed to carry modules and equipment to the surface, then return to orbit for refueling, with a minimal crew compartment. That vehicle you showed would be useful for carrying stuff around on the moon. I don't know if I'd go quite that far for routine flights too and from lunar orbit.

Unless harvestable lunar ice is actually found, the propulsion systems for both the lander and the tug would have to use hypergolic propellants. I'd keep the lander's systems as simple as possible and use pressurized propellant tanks. While this would result in heavier tanks, it'd eliminate turbopumps (a high maintenance item) and make throttling the engines much simplier. That's the approach used on the Apollo Service and Lunar Modules. I'd like to find a way to use more than one engine on the lander and arrange them in such a way that you could still safely fly the thing if an engine failed. You might have to jettison the external load but the goal is to allow either a safe landing or even better, an abort back to lunar orbit. This would probably require mounting the engines pretty close together and putting thrusters on outriggers to compensate for assymetric thrust.

The design requirements would call for simplicity, reusability, flexibility, and efficiency.