One of the objectives of spectrometers onboard space missions is to retrieve atmospheric parameters (notably density, composition and temperature). To fulfill this objective, comparisons between observations and model results are necessary. Knowledge of these model uncertainties is therefore necessary, although usually not considered, to estimate the accuracy in planetary upper atmosphere remote sensing of these parameters. In Part I of this study, “Computing uncertainties in ionosphere-airglow models: I. Electron flux and species production uncertainties for Mars” (Gronoff et al., 2012), we presented the uncertainties in the production of excited states and ionized species from photon and electron impacts, computed with a Monte-Carlo approach, and we applied this technique to the Martian upper atmosphere. In the present paper, we present the results of propagation of these production errors to the main UV emissions and the study of other sources of uncertainties. As an example, we studied several aspects of the model uncertainties in the thermosphere of Mars, and especially the O(1S) green line (557.7 nm, with its equivalent, the trans-auroral line at 297.2 nm), the Cameron bands CO(a3Π), and CO2+(B2Σu+) doublet emissions. We first show that the excited species at the origin of these emissions are mainly produced by electron and photon impact. We demonstrate that it is possible to reduce the computation time by decoupling the different sources of uncertainties; moreover, we show that emission uncertainties can be large (>30%) because of the strong sensitivity to the production uncertainties. Our study demonstrates that uncertainty calculations are a crucial step prior to performing remote sensing in the atmosphere of Mars and the other planets and can be used as a guide to subsequent adjustments of cross sections based on aeronomical observations. Finally, we compare the simulations with observations from the SPICAM spectrometer on the Mars Express spacecraft. The production of excited species at the origin of the green line, the CO Cameron bands and the CO2+(B) doublet is found to be on the dayside, consistent with photon and electron impact on CO2 as the main source of excitation of the three emissions, in contrast to the findings of Huestis et al. (2010) for the O(1S) case. Moreover, we re-examine the cross section for the production of the Cameron bands by electron impact on CO2.