Do you see many hollow airplanes? Wind around is the only way... wind 'through' still has to hit 'stuff', 'stuff' causes drag. That's why despite the number of piston engined small aircraft, few have liquid cooling systems, and most of the liquid cooled stuff winds up breaking even compared to air cooled - the added drag offsets the added power.
That said, you can reduce the speed of the air through the engine compartment or under the vehicle. Air dams, and divergent radiator ducts (like used on late model stock cars) can help.
The equation for aerodynamic drag is Cd*Q*A. Where Cd is the coefficient of drag, Q is the dynamic pressure, and A is the cross sectional area. Reduce any of them, you reduce drag. Q is 1/2*rho*V^2, where rho is density. So Q goes up with the square of speed. Then there's induced drag, which is drag caused by creating lift, or downforce. This is also linearly related to dynamic pressure, which is related to the square of velocity. So, this is why reducing the speed of the air through the innards of a vehicle can help - and with enough speed, you can still keep up with the requisite mass flow required to keep the engine cooled off.
Regarding shape: subsonic should be blunt end forward (think small airplane), supersonic you want pointy end forward (think Sr71). Dynamic pressure is highest at the stagnation point - or a surface which is completely perpendicular to the direction of flow. This is why a truck or van will have more drag than a something with the face slanted back some. But, that really only counts in clean air. The turbulence caused at the back is a whole other issue. Pointy end to the rear applies here (subsonic) and can have much greater gains than rounding off the front end.
If we were to take a fake 'Cd' number of say .50 for a truck, with a frontal area of 42 sq ft (7' wide, 6' tall - assumed), and a standard sea level density of 0.0765lbs/cubic foot, you'd have about 134lbs of drag at 50mph, 537 lbs at 100 mph, and 2150lbs at 200mph.
Lets look at 537lbs of drag at 100mph. With a 30" tall tire, that's a radius of 1.25 ft - which would require 430ft-lbs of torque in order to create 537lbs of thrust at the road surface. Assuming 1:1 at the trans, and let's say 4.10 at the rear end that's 105 ft-lbs of engine torque (110 ft-lbs with 3.91's).
You can find online calculators to play with and see what you can see. Obviously, the effect will be most pronounced with higher trap speeds, and the higher your average speed is. At the top end of the track, many drag cars aren't adding a lot more MPH, and usually because aerodynamic drag is becoming a bigger and bigger factor.
You choices to reduce drag are streamlining, and reducing frontal area. Both add up to lower your Cd. Lower roofline, smoother flow, less wetted area (which includes ANYTHING in the airstream - even under the vehicle if air flows there). Air dams, diffusers, louvers, etc can all help. Even turbulators on the surface can help energize flow and keep it from separating at areas near your mirrors, fender flares, side marker lights, etc.