Urban areas in the arid and semi-arid United States Southwest are facing increasing exposure to extreme heat, with compounding risks from both dry and humid heat extremes during the North American Monsoon (NAM). This study investigates how Gulf of California (GoC) moisture surges modulate the thermal performance of urban cooling strategies in the Las Vegas and Phoenix metropolitan areas using season-long, convection-permitting simulations at 900 m resolution. The model integrates two mitigation scenarios: combined cool roofs, walls, and pavements (CRWP) and a high-density street tree configuration (Tree-Max). These simulations confirm previous findings that daytime cooling is primarily driven by direct mechanisms, with high-albedo surfaces reducing solar absorption and vegetation enhancing shade and evapotranspiration, while nighttime cooling reflects indirect effects of reduced heat storage and radiative loss. The novel contribution of this study lies in explicitly linking these mechanisms to monsoon-driven moisture variability by compositing results under contrasting dry (no-surge) and moist (surge) regimes. This regime-based framework reveals that CRWP consistently delivers the greatest cooling across all conditions, whereas Tree-Max exhibits more localized and moisture-sensitive responses, highlighting the dynamic interplay between urban mitigation strategies and atmospheric moisture regimes. Nocturnal cooling enhancement is most pronounced under dry regimes, when efficient longwave radiative loss amplifies surface cooling. In contrast, during surges, elevated humidity and low- to mid-level cloud cover suppress radiative cooling and diminish the relative benefits of both tested cooling strategies. The findings highlight a strong radiative-convective coupling between urban surface modifications and the moist layers and clouds, exerting a key control on nighttime heat retention. These results demonstrate that the cooling effectiveness of urban mitigation strategies is not static but dynamically linked to monsoon-driven moisture variability.