Purpose: This research evaluated the relative effectiveness of three instructional approaches to supplementing the regular reading program for second graders with low word reading and/or pseudoword reading skills. Method: In the instructional experiment, 96 second graders with low reading achievement were randomly assigned to one of four conditions: (a) explicit and reflective word recognition, (b) explicit and reflective reading comprehension, (c) combined explicit word recognition and explicit reading comprehension, or (d) treated control that only practiced reading skills without any instruction. In the extension study, these conditions were compared to an untreated control group of 29 second graders. Results: In the instructional experiment, combined word recognition and reading comprehension treatment increased phonological decoding (pronouncing pseudowords) significantly more than the treated control or word recognition only treatment and had the highest effect size. The comprehension only treatment was not significantly different from the treated control. In the extension study, (a) the treated children receiving supplemental instruction improved significantly more in phonological decoding and reading real words than did those in the regular program, and (b) the combined word recognition and reading comprehension treatment, which was explicit, had the highest effect sizes for both pseudoword and real-word reading. Clinical Implications: The most effective supplemental instruction for increasing phonological decoding was combining explicit word recognition and explicit reading comprehension training.