A comprehensive uncertainty analysis was conducted on a recently proposed optimized detailed methanol/NOx combustion mechanism, focusing on its predictions for the experimental data used as targets in the optimization. The primary sources of model uncertainty in both the initial and the optimized mechanisms were compared. The propagation of uncertainties in kinetic and thermodynamic model parameters to the simulation results was investigated using approaches of varying complexity. Both absolute model uncertainties and, as a novelty, those normalized by experimental uncertainties were considered. The effect of correlation among the Arrhenius parameters in optimized reactions was examined through local uncertainty analysis. Accounting for parameter correlations yielded a more accurate representation of model uncertainty, although using the temperature-average of the uncertainty parameters also provided a reasonable approximation. The impact of correlations among all kinetic parameters was assessed using global uncertainty analysis with Monte Carlo sampling, which supported these conclusions. The analyses demonstrated that parameter optimization can significantly reduce model uncertainty. On average, the root-mean-square model uncertainty, normalized by the experimental uncertainty, decreased from a factor of 5.5 to 2.4 upon optimization. The dominant uncertainty contributions from the CH₂O + NO₂ = HONO +HCO and CH₃OH + NO₂ = HONO + CH₂OH reactions were effectively eliminated in the process. However, reactions involving the CH₂OH radical with NO₂, NO, HNO, and O₂ remained significant sources of uncertainty. To further reduce the model uncertainty, future research should focus on these reactions. This includes indirect experimental measurements sensitive to these pathways, as well as direct measurements or theoretical calculations of their rate coefficients.