The main objective of this study was to identify the probability distributions that best fit ozone (O₃) and nitrogen dioxide (NO₂) concentrations in Campo Grande, Brazil, considering three distinct periods: pre-COVID-19 (2005–2019), COVID-19 (2020), and post-COVID-19 (2021–2025). Monthly concentrations were analyzed using Lognormal, Gamma, Weibull, Log-Logistic, and Gumbel probability distributions. Model performance was evaluated using graphical analyses and the Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), Kolmogorov–Smirnov (K–S), and Anderson–Darling (A–D) statistics. The extension of the dataset through 2025 provided a broader temporal basis for assessing whether the patterns observed during the pandemic persisted in subsequent years. Mean O₃ concentrations were 260.17 DU, 265.87 DU, and 278.08 DU during the pre-COVID-19, COVID-19, and post-COVID-19 periods, respectively, whereas mean NO₂ concentrations were 2.97 × 1015, 3.11 × 1015, and 3.20 × 1015 molecules/cm2, respectively. The best-fitting probability distribution varied among periods and pollutants. During the post-COVID-19 period, the Weibull distribution showed the strongest overall support for both O₃ and NO₂. Overall, the extended analysis indicates that post-COVID-19 pollutant concentrations did not simply return to pre-pandemic mean levels. However, these temporal differences should not be attributed exclusively to the COVID-19 pandemic because meteorological variability, biomass burning, traffic, industrial emissions, and other factors may also influence atmospheric pollutant concentrations.