Drag coefficient comparisons are made for two half-size truck models with variable length, grain-haul height, closed trailers, and with identical frontal areas and length/volume distributions.One, a simple block model, with sharp edges, but characterized by correct overall dimensions, was designed to simulate only the major physical features of the truck, such as wheels, fenders, hood, cab and trailer. The second model accurately replicated body shape and hardware exposed to external and cooling air streams.Testing was conducted in the 9m x 9m NRCC, solid- wall wind tunnel of the National Research Council Canada (NRCC) over a velocity range of 48 to 193 kph (30 to 120 mph). Both models were yawed through ±14° with full-length and truncated trailer bodies.The measured coefficient data were corrected by adjusting the tunnel dynamic pressure, yaw angle, and the horizontal buoyancy effect using a simple modified pressure-signature correction method. This method was previously shown to result in correlating the wind-averaged drag coefficient data of the half-scale detailed model and actual full-scale vehicle within a 1% to 2% margin for the 7 mph wind speed and 55 mph road speed (7 and 55 mph) and 7 and 30 mph conditions, respectively.The half-scale model, representing a 2.2% frontal-area blockage in the wind tunnel was observed to be independent of Reynolds number effects inside the range of 2.2 x 106 to 4.4 x 106. Increasing trailer body length increased drag significantly. Engine cooling airflow was found to be the source of 3.8% of the total wind-averaged drag coefficient at a Reynolds number of 2.2 x 106, the ram and fan airflow contributions being 66% and 34% respectively. The difference in the wind-averaged drag due to model form and detail was inversely proportional to body length: a minimum of 2% for the full-length model and up to 11% for a 38% shortened model with the block model having the higher drag.