AIM: This study aims to evaluate the accuracy of scanned images of 4 clinically used intraoral scanners (CS3600, i500, Trios3, Omnicam) when scanning the surface of full arch models with various kinds of orthodontic brackets in the presence of artificial saliva. Materials and Methods. Four study models were prepared; bonded with ceramic, metal, and resin brackets, respectively, and without brackets. Reference images were taken by scanning the models with an industrial scanner. Study models were then applied with an artificial saliva and scanned 10 times, respectively, with the above 4 intraoral scanners. All images were converted to STL file format and analyzed with 3D analysis software. By superimposing with the reference images, mean maximum discrepancy values and mean discrepancy values were collected and compared. For statistical analysis, two-way ANOVA was used. RESULTS: Omnicam (1.247 ± 0.255) showed higher mean maximum discrepancy values. CS3600 (0.758 ± 0.170), Trios3 (0.854 ± 0.166), and i500 (0.975 ± 0.172) performed relatively favourably. Resin (1.119 ± 0.255) and metal (1.086 ± 0.132) brackets showed higher mean maximum discrepancy values. Nonbracket (0.776 ± 0.250) and ceramic bracket (0.853 ± 0.269) models generally showed lower mean maximum discrepancy values in studied scanners. In mean discrepancy values, the difference between scanners was not statistically significant whereas among brackets, resin bracketed models (0.093 ± 0.142) showed the highest value. CONCLUSION: Intraoral scanners and brackets had significant influences on the scanned images with application of artificial saliva on the study models. It may be expected to have similar outcomes in an intraoral environment. Some data showed the discrepancy values up to about 1.5 mm that would require more caution in using intraoral scanners for production of detailed appliances and records.