That is a considerable amount of variation, spanning over 2.5 SD, with certain clusters diverse enough to make the cluster score less representative of her true ability.
The Verbal Ability cluster is a combination of verbal comp and general info, which encompass an 18 pt range, over an SD. I would be concerned that expectations of MG-ness are an underestimate of her verbal reasoning ability.
Thinking Ability is derived from four smaller clusters: long-term retrieval, visual-spatial thinking, auditory processing, and fluid reasoning. (The component subtests are listed in pairs in your post, beginning with retrieval fluency.) FR has more weight in the TA score than the other areas do, which is why the TA score is not a simple average. There is some diversity among these subtests, but not a huge amount, as, though it spans just over an SD across the whole TA cluster, each of the clusters is relatively consistent internally. Long-term retrieval and visual-spatial are the weakest, in the high average range, and fluid reasoning the strongest, on the border of superior and very superior.
Cognitive Efficiency includes visual matching, decision speed, numbers reversed, memory for words (the processing speed and short-term memory clusters, respectively; I've listed them in pairs here). The range of subtest scores is relatively tight.
Where we get into some differences is in the Cognitive Fluency cluster (which was not reported), consisting of retrieval fluency, decision speed, and rapid picture naming. This cluster includes two of the lower subtest scores, and one of the higher ones, resulting in a range approaching 2 SDs. For her lowest, rpn, she was required to track across multiple lines on a page, naming images of familiar objects. This is not unlike what fluent readers do, once they are word calling, and not phonetically decoding every word. In rf, she listed as many words as she could in the allowed time, given a specific category. Usually kids with deep vocabularies (see her verbal comp score) do fairly well on this. We know from other data that the words are there, so this score suggests that her access to them is the gating item. She did well on decision speed, which does not require much in terms of retrieval skills, since the images are all provided. This task is different from the other two, in that it emphasizes reasoning (visual associations), rather than retrieval.
The pattern across this cluster suggests to me that retrieval efficiency in general is a vulnerability, and especially so when crossing visual and verbal modes, which has implications for reading. Though I am less concerned about visual tracking after seeing the strong scores in visual matching and decision speed, how much of a factor it is is still unclear, so that's definitely a rule-out I would pursue.
If you choose to pursue additional assessment, you may wish to consider measures of reading fluency (though you may have to wait a bit for those, as many of them are not normed for 5 yos, since most 5 yos are not reading; the PAL-II does include kindergarten norms). And, though the phonological processing tasks administered look pretty good, there are other aspects of PP that could be questionable, (specifically rapid naming) which might show up more strongly on a more comprehensive measure of PP, such as the CTOPP-2. (You have data that show this already, but in the event of an institution needing additional documentation, that would be one place to turn.)
