The first sentence in the first pair would most likely be interpreted as an unreal conditional about either the speaker or someone else, as you intended.
If we/you/they were living in Tokyo now, we/you/they would be caught up in the coronavirus pandemic.
In the situation of the second sentence, on the other hand, the speaker is necessarily talking about someone other than himself (or the listener). He doesn’t know for sure whether that person is living in Tokyo. She might be. If she is indeed living in Tokyo, she must be caught up in the pandemic.
Supposing they are living in Tokyo now, they must be caught up in the coronavirus pandemic (, but I’m not sure).
The uncertainty in the last part is due to んだけどね, not because of としたら.
Actually, this interpretation is not completely impossible from the first sentence, though it’s not very likely. としたら makes sure the sentence is understood as a supposition about something the speaker is not sure about.
The same logic works for the second pair. The first sentence is a simple conditional.
If Yamada-san hasn’t come, we cannot lift this table.
In the situation of the second sentence, the speaker is somewhere other than where Yamada-san is expected and is not sure whether he has reached there yet. Yet that place has to be one for which the use of 来る is logical and the table is where the speaker is (because of この).
Supposing Yamada-san hasn’t arrived there, we cannot lift this table.
I don’t know how but the decision of whether or not to lift the table must be somehow influenced by the presence or absence of Yamada-san in some other place.