The 84% and 79% in that image actually don't refer to the percent of questions you got correct. They refer to your
percentile in those areas, which is the percent of test-takers that performed worse than you in those areas.
By 'performed worse', they don't mean that those test-takers got fewer questions correct, either. The GMAT is scored in a different way from a high school or college exam. You can get a high score on the GMAT by getting relatively few questions right (I once got a 42 in Quant by intentionally missing every other question after the first five). You can also get a low score on the GMAT even if you get a lot of questions right (for instance, by missing a lot of questions in a row, or getting cut off at the end). Here are some more details on the algorithm - the article refers to Quant, but it applies to Verbal as well.
https://www.manhattanprep.com/gmat/blog ... mat-quant/The difference among your SC, RC, CR, and overall percentiles looks pretty extreme, but we have to assume they're correct, because GMATPrep is the only information we have (outside of taking an actual test) on how the GMAT scoring algorithm works. Here, it would depend on where in the test you missed a lot of RC questions, how difficult those questions were, and where you missed the SC and CR that you did miss.
One thing to note is that as far as we know, the test treats CR, SC, and RC identically. It's not like your score is calculated by first determining your 'CR score' and your 'SC score', and then using those to create your Verbal score. Instead, it looks at where you ended up at the end of the test, regardless of what you did with particular types of questions along the way.