<p>Urbanization has intensified concerns about climate and increased the demand for outdoor thermal comfort (OTC). The Comfort Formula model (COMFA) is an energy-balance-based approach that estimates human heat gain and loss from key microclimatic and physiological inputs. However, very few studies have examined the applicability of the COMFA model in composite landscapes, defined here as urban settings where buildings, hard paving, and vegetation jointly create strong short-distance heterogeneity in radiative and convective exposures (e.g., abrupt shading and sky-view-factor variability, facade-related longwave exchange, reflected shortwave load, and localized airflow changes). Here, we evaluated the applicability of the COMFA model in the hot and humid city of Guangzhou. To this end, we selected one simple configuration (a single-tree setting, Tree) and two composite landscapes (a high-rise plaza, High-rise plaza and a shaded high-rise plaza, Shaded high-rise plaza) in Guangzhou. Field measurements and thermal comfort questionnaires were conducted to obtain the microclimate characteristics and human energy budget. By examining 388 questionnaire records collected from 12 volunteer graduate-student participants over two summer monitoring days, a significant linear correlation (R² = 0.79) was found between COMFA and thermal sensation vote (TSV).Compared with the standard effective temperature (SET*), COMFA showed higher prediction accuracy across all three landscapes. While COMFA’s performance was slightly inferior to that of the physiological equivalent temperature (PET) in simpler urban landscapes, its accuracy improved with increasing landscape complexity. Thus, it outperformed the PET in shaded high-rise plazas. This study provides a preliminary exploration of the applicability of the COMFA model in urban environments. To strengthen the reliability and generalizability of the findings, future research will aim to increase the sample size and further validate the model’s applicability.</p>