This study is based on previous works in a series of numerical studies on submarine hydrodynamics, which involved developing a computational fluid dynamics method to estimate the self-propulsive performance of underwater vehicles. Herein, the Defense Advanced Research Projects Agency SUBOFF submarine model was adopted as a benchmark. The computational modeling applied was based on the Reynolds-averaged Navier-Stokes turbulence model. A body-force propeller method was adopted to model the propulsion. The self-propulsive performance was verified via mesh refinement and validated by comparing the computational solutions with the results obtained from the experiments. The effect of the Reynolds number on the self-propulsive performance was investigated by varying the positions of the stern planes, while the free surface effect was determined by varying the Froude number (Fr) via the volume of fluid method. The computed Taylor wake fraction (w) and hull efficiency (ηH) depended on the Reynolds number as it decreased monotonically. The w and thrust deduction fraction (t) for the model of aft-fitted stern planes were approximately 3–7% and 8–10% higher than those of the baseline and fore-fitted stern planes, respectively. The differences in ηH between the models were insignificant. Regarding the free surface effects, the computations of w, t, and ηH generally decreased with Fr, thus exhibiting several humps and hollows. The computed upward suction force and pitching moment varied from negative to positive and vice versa, depending on Fr.