It’s not really a Boolean question of it being broken or not, but rather more on the cost of opportunity the current software represents in relation to having a much more customizable, secure and dynamic evolving software that could benefit the forum on all fronts (user, programming, admin/mods, [advertisers]).
New software is always very expensive to develop, given the high cost of developers/engineers.
For whatever reason, SMF appears to be fairly inflexible.
Of course, something’s clearly gone wrong along the way with the new software, likely time and over again, and I’m sure that, were it to be ready today for the migration, it would require a breath-taking amount of time to kick-off, supervise, correct and evolve over the following months, all of which require setting aside ample available time and resources (be them prime or delegated) to embark.
At the end of the day, a forum is a database, and forum software will query said database.
Part of a forum database is various permissions that each user has -- this can be simplified by creating user groups that a user is assigned to, and giving the user group various permissions, and flexibility can be added by adding multiple group types, such as 'native_rank, paid_group, and banned_group'. For example to check if a user has permission to do something, the output of MIN(banned_group, MAX(native_rank, paid_group)) could be used.
Forum software ultimately needs to be programmed to make various database queries, and present the output of said queries in a user-friendly format.
When you look at forum software as the above, your only limitations on features are your ability to create programmatic database queries.
I really don't think there is any reason to reinvent the wheel. I would propose creating a flask app for the new forum software. Adding new features would be as simple as editing the database schema (if necessary), writing a SQL query, and editing a little bit of HTML.