Fuel treatments are commonly applied to increase resilience to wildfire in dry and historically frequent-fire forests of western North America. The long-term effects of fuel treatments on forest structure, fuel profiles (amount and configuration of fuels), and potential wildfire behavior are not well known relative to short-term effects. Additionally, long-term treatment effects on the development of stand structure and fuel profiles have rarely been compared to the long-term effects of pre-treatment conditions, treatment intensity, and site productivity. In this study, we addressed these knowledge gaps by resurveying 204 permanent plots at the Northeastern Cascades site of the Fire and Fire Surrogates study 13–18 years (‘long-term’) after burn-only, thin-only, and thin plus burn treatments, and comparing results to pre-treatment conditions and un-treated controls. Methods included ordinations, generalized linear mixed models, and fire models. All treatments shifted long-term average stand structure toward lower tree density, basal area, and crown fire potential, with thin plus burn treatments showing the highest magnitude of effects for most variables. However, the direction and magnitude of trajectories among plots within treatment types were highly variable. Long-term responses of stand structure, fuel profiles, and modelled fire behavior were positively correlated with their pre-treatment values. Treatment intensity strongly affected long-term stand structure and canopy fuel loads. By ∼15 years post-treatment and under 80th percentile fire weather conditions, most plots in all treatments and un-treated controls failed to meet target thresholds for surface flame length, basal area mortality, and torching index, while most plots met thresholds for crowning index. However, live stand structure following wildfire simulated under 80th percentile fire weather conditions was characterized by lower stand density and a shift toward dominance by large-diameter and fire-resistant trees, suggesting that treated stands may be resilient to wildfires occurring under moderate weather. Our study suggests that understanding fuel treatment efficacy and longevity may be improved in future studies by incorporating fine-scale (i.e., plot-level) drivers of variability in stand structure, fuel profiles, and modelled fire behavior, and by using multiple methods of evaluating treatment effectiveness. Thin plus burn treatments and intensely applied burn-only and thin-only treatments can reduce basal area and potential for crown-fire for more than a decade. However, additional maintenance treatments may be needed by 15 years after initial treatment, to further reduce potential for severe surface fire and high tree mortality in subsequent wildfire.