Also ask if adding the randomized delay would be compatible with use of I2P as a foundational network layer for general network functions? Answer: No
Not sure I understand, this delay is useful when sending coins because couple of additional seconds is worth the wait. We don't need other general network functions - or do we?
I2P is a general network layer. It is not just Monero that is being run on it. Adding a protocol change to I2P impacts all the network services that run on it. I suppose it might be possible to allow services to opt-in higher latency, so that is implicit in the question I asked of them. But it might not be possible for various reasons, one being resource usage reciprocity on relay nodes (or who provides the bandwidth for free and why are they not the NSA's Sybil attack?).
To get to Visa scale with Bitcoin brings us to 80 Terabytes.
Monero claims 5 - 6x bloat (and I disputed that, it might be higher). So heading towards a Petabyte just to reach Visa scale.
But..
Do we expect and require Visa scale and micro transactions? Bitcoin is fine for that. Remember you yorself claimed we are running out of time because of coming global sovereign debt collapse and currently there is nothing better on the horizon that would stand up to your criticism

for now let's just stick with what we have and concentrate on the main goal - it doesn't include micro transactions on Visa scale.
If something comes along that can scale faster because supports ZERO transaction fees, prevents pools from being large, mass mining, micro transactions and Ethereum style contracts, then Monero could be surpassed.
Also I don't believe Monero is anonymous to the NSA and authorities with high reliability given the weaknesses in I2P and Tor.
Also just like Bitcoin, Monero appears headed towards centralization thus it won't take long for NSA to have power to send national security gag order to the few pools that will dominate it, and thus defeat the anonymity either by correlating entry and exit node timing or simply changing the protocol to ban transactions they can't correlate.
Centralization is the near-term enemy. Bitcoin is already there. Monero has not innovated at all on decentralization of mining.
Note I did not argue that Monero is useless in all possible scenarios one can contemplate.