Specifically, this experiment eliminates (much of) the problem that earning a few hundred euros will cut your unemployment benefits by the same amount. It allows people to take part-time jobs without being penalized. It's not perfect for the reasons you give and more, but it might be significantly better than what is currently in place.
The premise that one can find some universal income amount X and do away with all other types of benefits is a pipe dream anyway. For some, X will be enough to scrape by. For others, X will not be enough for their medical care necessary to survive for two weeks. There will always be a need for some extra needs-based benefits.
Finland (like much of Europe) has universal healthcare and a national health insurance plan; this is not an issue. There already is an 'X' that people can (adequately) survive on when on welfare. If basic income is something that a state wants to research in earnest, these kind of experiments provide valuable data and can be seen as a tentative first step towards a national basic income plan.
That is why this a useful basic income experiment - regardless of whether the program itself is classified as basic income in its current form or not.
Not really, because (i) tax credits and other proposed "welfare trap" reduction methods have been around and studied for a while and (ii) it doesn't attempt to study any of the features of BI widely believed to have negative impacts such as the cost of extending welfare to millions of people not currently [interested in being] eligible for it, possible social effects of decoupling benefit entitlement from any indication of desire/need and net income reductions to some current welfare beneficiaries if other programmes such as housing entitlements are cut. As a general rule, experiments which don't test any of the perceived negative effects of a proposal generally mislead more than they inform in debates about whether it's an improvement on the status quo.