For example: If the outgoing addresses include not only 13zb1hQbWVsc2S7ZTZnP2G4undNNpdh5so but also a new address 1KDwiguXGo4JFJoHA3azAHx4Neqkr76kKs with an outgoing amount of 10,000 satoshis, and the attacker does not possess the private key for 1KDwiguXGo4JFJoHA3azAHx4Neqkr76kKs, can the transaction originating from 13zb1hQbWVsc2S7ZTZnP2G4undNNpdh5so still be replaced?
Yes the transaction still can be replaced, as long you pay more fee it always will be accepted by the miners.
What is going to happen with the TX from 1KDwiguXGo4JFJoHA3azAHx4Neqkr76kKs ?, well as long the new RBF TX is accepted and actually mined, the outgoing TX from 1KDwiguXGo4JFJoHA3azAHx4Neqkr76kKs will be droped by the mempool.
I already did this test before on testnet and mainnet and the original TX was always replaced, the new TX was accepted and mined eventually. Old Outgoing TX was doped because the original TX was not longer valid.
I 2nd this - Albert0 is correct. I did this same test as well, and found that the mempool will still accept the higher-fee RBF transaction, no matter what. It does not matter if your transaction has RBF disabled, it will still be replaced by a higher fee transaction.
I even took it a step further - I tried an identical transaction as described as above, and then I immediately spent the transaction (call it transaction1) to a 3rd wallet (transaction 2), all while there were still 0 confirmations (I spent the unconfirmed coins from transaction 1 to transaction 2). So Then I submitted a replacement transaction for transaction 1 (which technically has been spent already) with a higher fee, and the mempool accepted it and discarded transaction 2. All these transactions were with RBF disabled. So really, the only thing that disabling the RBF option does, is it just prevents YOUR wallet from sending a RBF transaction, but it doesn't stop anyone else from doing so.