Clarify need for log_slave_updates in Galera to have reliable binlogs

Hello,

I would like to ask for clarification of topic, whether it is possible/safe to use binlogs from one Galera node to PITR another one, once proper position has been located in the remote binlog. Binlogs in question weren't created with log_slave_updates=ON on neither node. No GTID params are set, server_id is identical across all nodes.
The issue is, that all my attempts to locate entries which, based on how I understood documentation, should be missing on READ nodes, were actually present in binlogs across whole cluster. No matter if the node received or created writeset. With this experience, I find following text somewhat confusing: (link) -
"Like with MariaDB replication, write sets that are received by a node with Galera Cluster's certification-based replication are not written to the binary log by default."