The quality of local skin temperature prediction by thermophysiological models depends on the local skin blood flow (SBF) control functions. These equations were derived for low activity levels (0.8-1 met ) and mostly in sitting or supine position. This study validates and discusses the prediction of foot SBF during activities of 1-3 met in male and females, and the effect on the foot skin temperature prediction (∆ꓔ skin,foot ) using the thermophysiological simulation model ThermoSEM. The SBF at the foot was measured for ten male and ten female human subjects at baseline and during three activities (sitting, walking at 1 km/h, preferred walking around 3 km/h ). Additional measurements included the energy expenditure, local skin temperatures (ꓔ skin,loc ), environmental conditions and body composition. Measured, normalized foot SBF is 2-8 times higher than the simulated SBF during walking sessions. Also, SBF increases are significantly higher in females vs. males (preferred walking: versus 4.8 ±1.5 versus 2.7 ±1.4, P < 0.05). The quality of ∆ꓔ skin,foot using the simulated foot SBF is poor (median deviation is -4.8 C, maximum deviation is -6 C). Using the measured SBF in ThermoSEM results in an improved local skin temperature prediction (new maximum deviation is -3.3 C ). From these data a new SBF model was developed that includes the walking activity level and gender, and improves SBF prediction and ∆ꓔ skin,foot of the thermophysiological model. Accurate SBF and local skin temperature predictions are ben-eficial in optimizing thermal comfort simulations in the built environment, and might also be applied in sport science or patient's temperature management.