When can we say that a manuscript 'apparently' supports a reading?

Sometimes there is a certain amount of doubt what a manuscript reads at a particular point. If there is a variant reading and a manuscript has only a few of the letters but these letters fit with one reading and not with the other it is acceptable practice to cite this manuscript with the qualifier videtur ('apparently') in support of the reading that fits. A good example is P70 in Matthew 2:23 where the reading ναζαρα for ναζαρετ is accepted as the apparent reading of P70, even though only the final two letters are visible (I am trusting the transcriptions here). For all we know P70 could have read γαδηρα, but because of Eusebius and the comparable variant at 4:13 there is a good case to be made for ναζαρα in P70. I think that this example is more or less on the edge but fine as it stands.

But what about the following case in James 4:13?
πορευσόμεθα εἰς τήνδε τὴν πόλιν καὶ ποιήσομεν

There is a variant πορευσωμεθα and also ποιησωμεν. Most manuscripts have either twice the indicative or twice the subjunctive, an important few have first the indicative and then the subjunctive. There is none that has first the subjunctive followed by the indicative.

This is what P100 has:


αυρι]ον πορευσ[
]ποιησομεν[

As we can see, P100 reads the full ποιησομεν but in the line above we have only -ον πορευσ. I cannot see how anyone could argue for a following omicron over an omega. Still, the NA27/28 ECM1 all have P100 as πορευσομεθα ut videtur.

Is this justified? Yes, there is a case why in light of the following indicative ποιησομεν it is likely that P100 has also the indicative πορευσομεθα here (since there is no other manuscript with the subjunctive first and then the indicative). Or 'No', since this argument is only indirect and not based on any observation of letter shapes.
I am not sure about the correct answer, it just shows that it is important to check videtur whenever possible.