We develop a realistic and time‐varying model that satisfies both the Voyager 1 (V1) and Voyager 2 (V2) observed crossing times and locations of the termination shock (TS) simultaneously by performing three‐dimensional (3D) magnetohydrodynamic (MHD) simulations using a total variation diminishing (TVD) scheme that includes the effects of neutral particles. Daily values of solar‐wind speed and density observed by V2 are used at every simulation step so that short‐term dynamical effects are reproduced. Before performing the dynamic simulation, we generate a 3D stationary solution using a set of standard parameters: the interstellar proton density is assumed to be 0.061 cc−1, the interstellar neutral hydrogen density 0.176 cc−1, the interstellar medium speed relative to the Sun 26.3 km s−1 and a temperature of 6300 K. The interstellar magnetic field intensity is 0.3 nT and the orientation is oblique to the flow direction lying in the hydrogen deflection plane. The anisotropy of the solar‐wind speed is also taken into account using 400 km s−1 at low latitudes and 1.5 times faster at high latitudes. A solar‐wind density at 1 au of 3.55 cc−1 and a latitudinal angle separating the high/low solar‐wind speed regions of 80° yield dynamic solutions, starting from the stationary solution, that satisfy the V1 and V2 crossing times and locations. Our simulations clearly show that (i) the TS position increases whenever a solar‐wind high‐ram‐pressure pulse collides with the TS; (ii) a large‐amplitude magneto‐sonic pulse is generated downstream of the TS when a solar‐wind high‐ram‐pressure pulse collides with the TS; (iii) fine structure in the heliosheath is present, corresponding to a magnetic wall and plasma sheet near the heliopause (HP), and a thin current sheet; (iv) the magneto‐sonic pulse is reflected at the plasma sheet in the heliosheath, and the TS position decreases when the reflected pulse collides with the TS and (v) the time‐varying radial distance between V2 and simulated TS positions appears to be consistent with the TS particle intensity profile during the period when the TS particles were observed at V2. This last point, together with the overwhelming consistency of our MHD results with a standard set of solar‐wind and local interstellar medium (LISM) parameters, suggests that the outer heliosphere is controlled essentially by MHD and neutral interstellar gas processes, although V2 observations of solar‐wind plasma in the heliosheath are not well reproduced in our simulations.