I think PIPS is far more useful than cognitive ability testing, as long as it is followed up by some intervention for those children with deficiencies identified. Children who can be trained in the early years of school or even in nursery to use thinking skills and risk-benefit analysis to solve problems will usually avoid the sort of disruption issues that teachers find off-putting, may develop a more positive self image as their behaviour is rewarded in the classroom, and may become better able to gauge the relationship between cause and effect, thus engage more with the school learning process. I think socialisation is what schools should be concentrating on in the early years and that any strong academic focus is actually misplaced. I think good socialisation and the development of resilience (the ability to try new things and to bounce back from failure) would be more profitable educationally in the long term.
Cognitive ability testing relies on correct administration in an optimal environment in order for any meaningful data to be extracted. Physically, the child being tested needs to be able to hear accurately and understand the instructions in whatever language they are given. He or she needs to be able to see the symbols clearly. The child needs to be co-operative and wide awake, healthy and preferably fed that morning. Ideally, the child needs to be ready emotionally for the test and not distracted by traumatic events in his or her life (witnessing domestic violence, suffering violence, strain or worry about a family member's health, arrival of new baby, etc). In some tests, such as Woodcock Johnson III iirc, the child is given feedback in the process of a section of the test and depending on individual resilience (which is not the area being tested) might be put off by hearing that he or she had got an answer wrong. In addition, the sort of puzzles and tasks set for the subjects may be completely unfamiliar to some but familiar to others depending on what parents see fit to provide in the home environment; children who are thrown by the appearance of the test, the surroundings, the appearance or other personal feature of the test administrator (smelly, pretty, old and wrinkly, scary, etc.) or the tasks themselves may not feel comfortable enough to settle down and focus.
I think the best you can really say about results is that on that day, so-and-so performed well under the conditions that pertained. I think you need a wide range of assessment over time to see where a student really is. However, I don't think that assessing where a student is tells everything a teacher should know about where a student could be, given sufficient effort by all parties to come up with an appropriate teaching and learning strategy for that individual.
The science of cognitive ability testing is not value free and conclusions may not hold for all cultures or even for all personalities (the nature vs. nurture debate rears its head here). I think tests are most useful in identifying children on the upper and lower ends of the scale -- those who may be considered G&T or those who have serious intellectual challenges. In the very young, it's difficult to discriminate what lies behind a low score however. A zero due to stroppiness looks exactly the same as a zero due to incomprehension. They don't really say much about the potential of the middle, and actually they don't really say much about the potential of the G&T crowd either, since realisation of potential often depends on emotional, psychological and even cultural and economic factors.