This is in Gregory & Tuckett's The New Testament and the Apostolic Fathers: The Reception of the New Testament in the Apostolic Fathers (amazon.com), p. 130, note 5:

Hagner's monograph contains an extended methodological discussion of how scholars should evaluate what he refers to as 'variant [i.e. inexact] quotations' (in which he argues that these are usually best explained as memoriter quotations form known texts rather than as accurate quotations from unknown texts or oral traditions; see Hagner, Use [of the Old and New Testaments in Clement of Rome], 80-108, on the use of the OT; 287-312, on the use of the NT) and also a helpful survey of how the pattern of 1 Clement's apparent use of the writings later canonized as the NT compares with that of the use of the same writings in other Apostolic Fathers (ibid. 272-87).

So, Hagner's distinction and methodology is helpful. Basically, unattested quotations/allusions are better treated as inexact quotations of known readings than as exact quotations of unknown readings.

I suppose this appeals to the lowest-common-denominator and is right more often than not, but of course it also means that in practice, all quotations/allusions are always accounted for among the known readings, so, for text-critical purposes, the Apostolic Fathers are ultimately useless as regards attestation of heretofore unknown readings.

In practice, however, this also means that where there is similarity between a quotation/allusion in the Apostolic Fathers, it can be used to help understand how the writer understood the NT text without falling into a text-critical black hole. And, at least for my purposes at present, this is helpful.