ABSTRACT A previously developed predictive model for the decimal reduction times (D72°C) of Salmonella typhimurium (ATCC 13311) in simulated citrus systems was validated for its efficacy to estimate the D72°C of the reference organism in real citrus juices. Model predictive efficacy was assessed by calculating performance indices for accuracy (Af) and bias (Bf); and through graphical validations. In all validating citrus juices, the predicted D72°C of the reference strain were greater than the actual values calculated from thermal inactivation studies. The model overestimations were also demonstrated in the graphical validations where all model predictions were classified as fail-safe predictions. Results obtained in this study may be used in establishing new processes as well as in the verification of existing processes for citrus juices to ensure the safety of the products against the reference organism. PRACTICAL APPLICATIONS This study dealt with the validation of the predictive efficacy of a previously developed regression model for the thermal death times of the pathogen Salmonella typhimurium in citrus juices. The validated model may be used in estimating the thermal death times of the reference pathogen in appropriate citrus juices and applied in establishing thermal process schedules. Recognizing that the more traditional means of establishing the thermal death times of a specific organism in a specific food system entails rigorous work, the use of the validated model may be considered as a convenient alternative.