that's a complex question. when you "clean" the head to the point that you see that pretty fine parallel spray coming out, many times, in order to achieve that spray, the head is being damaged. so you have clogged head, cleaned out by high pressure purges, now that it's cleaned out it's sustained mechanical damage. i've messes up quite a few heads cleaning them. and in retrospect if i'd stopped short of wanting to see that stream so perfect, i think i would have had a working head in the end.
the head is full of soapy water. glycol ethers and acetone are both miscible in water. i like to use straight carbitol (one could use cleaning solution). i don't know why there is a special fluid designated for purging. i figure it's the same stuff with possibly more humectant to bridge and pull out the most possible water from every tiniest spot. in general if you get a perfect stream, you should be getting no deflected or missing nozzles... that is if they can fire properly after being over pressured.
one piece of advice i received once is when cleaning out waterbased heads to a little stronger with the pressure and go much longer. this goes against my theory that the water based and the solvent based head has the same solution in them. but it's what someone else believes to be true.
the data issue doesn't make sense. maybe something will come up to explain it other than a faulty head. but nothing so far.