Video self-modeling (VSM) has been used to improve the social behaviors of students from diverse backgrounds. However, its application to reading fluency is still in the preliminary stage. VSM can be divided into feedforward VSM (FFVSM) and positive self-review (PSR). Previous studies examined the effect of stand-alone FFVSM and its effect when adding it to intervention packages or repeated reading (RR). However, no study has examined the effect of PSR on improving reading fluency. Following a baseline phase, an alternating treatments design compared the control, PSR, RR, and PSR + RR conditions. The comparisons evaluated the stand-alone effect of PSR and the effect of PSR as an intervention component when added to RR. The extended analysis and follow-up phase validated the most effective intervention identified in the alternating treatments phase for each participant and evaluated the long-term effect of the most effective interventions. Based on visual analysis and two effect size measurements (Nonoverlaps of All Pairs and Tau-U), PSR was found to be effective for only one of three elementary school students with disabilities. Moreover, RR was more effective than the stand-alone PSR for the participants. Practical implications, limitations, and future studies are discussed.