>Surdéclinaison indeed seems at least to correlate with group
>inflection, while suffixaufnahme is a matter of word inflection

An addendum regarding that and Early Qenya: In the mentioned source, Tolkien
actually lists nominative, accusative, genitive, and dative as 'cases', but
inessive _-sse_, ablative _-llo_, allative _-nta, -tta_, partitive _-inen_,
manner _-ndon_ as 'adverbial suffixes'. About the differences between them
he writes:

"(i) They (the adv. suffixes) naturally cannot be all formed from every noun
and adjective; (ii) they are never added, except in verse, to an adjective
in agreement with an expressed noun: where a qualified noun receives one of
these endings (a somewhat archaic mode) the adjective usually precedes
uninflected (except rarely for plural) and is virtually a loose compound"

In other words, the cases are word-marking with agreement:
_ni heps-ine [...] i mailin-e-n losse-li-n_ 'I bound [...] the beautiful
I bind-PST ART beautiful-PL-ACC flower-PL-ACC

The adv. suffixes are group-marking:
_taara kas-i-sse(-n)_ 'on the high tops'
high head-PL-INE(-PL)

And it's the adv. suffixes which show a certain degree superflection.

(It also seems that the 'cases' show signs of fusionality, e.g. acc. sg.
_-t_ is different from acc. pl. _-lin_, while the adv. suffixes always
agglutinate the plural marker to the singular form.)