Far transfer is rare to occur, and a recent meta-analysis suggests that music is no exception. The overall effect of musical training on cognition was claimed to be null when considering studies with active control groups or implemented randomization procedures (Sala & Gobet, 2020). Using the authors’ data file and program (https://osf.io/rquye), we did not confirm the effect of randomization, and we demonstrated that their conclusion is based in part on the failure to differentiate near and far transfer, with near transfer effect sizes being selectively excluded for the musical training group studies, but not for the active control group studies. Reanalyzing their data file resulted in a significant effect size (g = .234), and also provided new evidence that far-transfer effects of musical training can challenge near-transfer effects of linguistic training. Music is a recreational activity that may be special in allowing for small, but statistically significant far-transfer effects.