Neither of your formulations quite fits what appears to be the 'standard' analysis of this sort of construction.
These days the infinitival clauses, including their subjects, are usually characterized as non-finite complements of the main-clause verbs (not objects of the verbs, because that term is restricted to noun phrases, including pronouns, which are actually pro-NPs). Some grammarians maintain that such infinitival clauses are introduced by the complementizer/subordinator, for, which may be omitted in some contexts, just like the complementizer/subordinator that which introduces finite complement clauses ('content clauses').
Pronoun subjects of infinitival clauses are cast in object case, but they are not actual objects of the main-clause verb (or of for, if that is present): the subject I of the main clause doesn’t want or need ‘her’ herself or ‘you’ yourself, what I wants/needs is for the action to be performed by ‘her’ or ‘you’.
These pronouns are rather said to be “raised to object”; this is a metaphor drawn from the structure of syntax trees, where the subject ‘moves upward’ to the object slot of the main-clause verb. You might think of them as objects-by-position as opposed to objects-in-fact.
Man_from_India points out that these sentences, without the context of your question, are ambiguous. The subjects of the infinitivals may not be the pronouns her and you but the subjects I of the main clauses:
I wanted her in order that I might do it.
I need you in order that I may do it.
If this is the case then the pronouns are indeed objects of the main-clause verbs, and the inifinitivals are indeed "adverbial" adjuncts (not complements) modifying the main clauses—they can be moved to the front of their sentences without changing the meaning:
To do it I wanted her.
To do it I need you.
The ambiguity can only be resolved by the discourse context.