An unexpected "J" aggregate can form at any time. For example, a dye that is a green sensitizer for a Bromide emulsion can become a red sensitizer for a Bromo-Iodide emulsion due to strong adsorption. The same can happen for almost any dye used improperly.
BTW, the dye I refer to above is actually used as a red sensitizer in a high iodide Kodak product, so this is no useless or rare phenomenon, but when it hits you, you get unexpected and "wrong" results. For example, expecting green and getting red sensitization.
Therefore, it is possible to imagine a short IR dye becoming a long IR dye if a "J" aggregate forms or a red sensitzer becoming a short IR sensitzer if a "J" aggregate is formed. There is no way to really tell beforehand without testing on the emulsion in question.
Regarding erythrosine, it is usable and gives fine results if you know how to use it. If you don't know how to use it properly it will give poor results. It is also quite economical, as it is about 1/4th the price of many other common sensitizing dyes and is more stable. It is therefore useful for doing a lot of early on experiments without eating up a lot of expensive dye.
Also, the expensive dyes are hard to get, and becoming harder to get all the time.