Physiological modeling is important to evaluate the effects of heat and cold conditions on people's thermal comfort and health. Experimental studies have found that older people (above 65 year old) undergo age-related weakening changes in their physiology and thermoregulatory activities, which makes them more vulnerable to heat or cold exposure than average aged young adults. However, addressing the age-related changes by modeling has been challenging due to their wide variability among the older population. This study develops a two-node physiological model to predict the thermal response of older people. The model is built on a newly developed two-node model for average-age young adults by accounting for the age-related attenuation of thermoregulation and sensory delays in triggering thermoregulatory actions. A numerical optimization method is developed to compute the model parameter values based on selected benchmark data from the literature. The proposed model is further validated with published measurement data covering large input ranges. The model predictions are in good agreement with the measurements in hot and cold exposure conditions with a discrepancy 0.60 ℃ for the mean skin temperature and of 0.30 ℃ for the core temperature. The proposed model can be integrated into building simulation tools to predict heat and cold stress levels and the associated thermal comfort for older people in built environments.