Did the neuropsych explain what the basis for the Anxiety NOS Dx was? Or comment on its possible impact on any of the cognitive testing? Anxiety can affect any subtest, but I definitely find that it more often affects the WMI subtests. Also, the block design score may have been affected by anxiety as well, as that is the first subtest administered, and is also timed. Sometimes anxious kids have a hard time settling into the testing situation, especially with a somewhat unfamiliar task type. (It's first because it's supposed to be engaging and non-academic, but that doesn't work out for every kid.)
He has very good verbal skills, a definite relative strength, in what I would consider the gifted range on this instrument, accompanied by excellent and comparable processing speed. These two things put together usually result in high school achievement, especially in the primary and elementary years, when even mathematics is really more of a language task.
If the PRI scores are all described as valid (including bd), I would investigate a nonverbal learning disability (NLD or NVLD). It is not unusual for anxiety to accompany this learning profile, as rules, routines, and meaningful contexts are particularly important for individuals with NVLD. Many are misdiagnosed with Asperger's/HFA, or have it in the differential Dx for a long time (the field is still discussing whether they are truly distinct Dx, but most currently agree that HFA usually have both high visual-spatial and poor social communication, where NVLD may have the social deficits, but will have visual-spatial weaknesses, too). ADHD is also a common confound.
Typically, I would not see this high of a PSI in NVLD, but, in this case, it might be explained by the difference between rote visual-motor tasks (one is not even a significant fine-motor task, while the other is very loosely scored for fine-motor accuracy) and visual-spatial problem-solving tasks. Notably, the PRI task which he did the best on is somewhat amenable to verbal mediation.
On memory, in addition to anxiety, there is the question of how he did on digits forwards vs backwards, which are not exactly the same cognitive task. If, say, backwards were much weaker than forwards, that might support a visual-spatial weakness. If the reverse, I would lean more towards anxiety as the issue, in this case, given the solid IVA attention scores.
