We continue the development of a modeling system for the investigation of lightning-produced NOx at the process level by including the chemistry aspects used in the companion paper <Zhang et al., this issue> (hereinafter referred to as Part 1) in the three-dimensional version of our Storm Electrification Model. Looking toward longer simulations, we expand the chemistry to include CO, CH4, OH, and HO2, with HNO3 as a sink. As in Part 1, we simulate the 19 July 1981 CCOPE cloud using noninductive charging for electrification, producing 18 intracloud lightning flashes over a 3-min period. The simulation ends at 38 min with the cloud dissipating. Energy dissipation due to lightning in this simulation ranges between 0.91 and 2.28 GJ. Results show a maximum NO mixing ratio of 35.8 ppbv produced by lightning during the simulation. As the cloud dissipates, following the cessation of lightning, there are maxima for both NO and NO2 of ~6.3 ppbv around 4 km altitude. The NO mixing ratio in the anvil peaks around 2 ppbv near 10.5 km. These results are in reasonable agreement with available observations. A striking feature is a plume of NO2 with mixing ratios of the order of 0.5 ppbv reaching the surface. There is no similar plume for NO. Likewise, NO from the core of the cloud is transported into the anvil, while NO2 does not exhibit the same behavior, probably as a result of photolysis. The NO2/NO ratio is found to decrease with altitude and is comparable to estimates derived from observations. The NO production per unit length (mean = 2.03 ¿ 1022 molecules m-1) is also within the range of estimated values. Our results indicate that short-lived storms may produce a vertical profile of NOx that differs from the C-shaped profiles of Pickering et al. <1998>.