Some studies have addressed the influence of optical magnification on the detection of caries lesions using a visual scoring system. However, there is a lack of research related to the use of the CAST and ADA-CCS visual scoring systems. In addition, the reliability and accuracy of ADA-CCS index in permanent teeth were not studied yet. So, the aim of this study was to evaluate, in vitro, the influence of different levels of optical magnification on the detection of occlusal caries lesions in permanent molars using three visual scoring systems. One occlusal site per tooth was analyzed in 120 extracted permanent molars. Two trained examiners inspected the teeth using ICDAS (International Caries Detection and Assessment System), CAST (Caries Assessment Spectrum and Treatment), and ADA-CCS (American Dental Association-Caries Classification System) visual criteria, twice with each scoring system, with a one-week interval between examinations. The study was conducted in three phases: (A) without optical magnification, (B) using a binocular lens (3.5× magnification), and (C) using an operating microscope (16× magnification). Then, the teeth were sectioned longitudinally through the center of the selected site and the section with the more severe lesion was histological evaluated considering the D1 (lesions in enamel and dentin) and D3 (dentin lesions) thresholds. Kappa values for intra- and inter-examiner reproducibility were good to excellent for all systems. At the D1 threshold, sensitivity, accuracy, and area under the ROC curve were high for ICDAS and CAST in all phases. However, this was not the case for the ADA-CCS in phase C (<0.05). At the D3 diagnostic threshold, there was no significant difference between the visual scoring systems during the study phases (>0.05). The magnification does not improve the accuracy of the visual scoring systems in the detection of occlusal caries lesions in permanent molars.