@L&LD I think it's a valid questions though. And it actually goes hand-in-hand with your point on DNS.
Fundamentally, the issue with security is, and has always been, that you have to decide who to trust.
Let elaborate a little bit as it seems relevant to the OPs question:
First, it's important to understand who the threat is we're trying to protect ourselves from. Right up front, if it's nation states, well we're done, don't have the resources or skills (for 99.99% of users most likely) to prevent anything. They have attack vectors (like threatening developers or finding folks inside organizations) to forcing companies to comply *ahem*China*ahem*.
So if focusing on the more criminal element, then some of those go away. So let's say we trust the folks that write software for our routers, we need to do that for both the vendors (Asus, partners like Trend who provide software / services and 3rd party developers, like RMerlin, Adamm etc).
Second, we already know that some of those folks are not trustworthy, like Trend Micro. They've had numerous issues in the past and I would personally not turn on anything that would submit a significant amount of my data to them. Others, like the folks writing 3rd party software, are
most likely trustworthy, but I don't know their background. I have no concerns but am simply expanding on the thread vector here.
Third, communicating across networks we will always expose information about ourselves to someone, though we get to decide who we expose what to using technology.
Ex. Who see what you are searching for on the web, the sites you visit and, to some degree, what you access on those sites.
DNS being the starting point. Without doing anything, our ISPs know exactly what we visit and how often (to a degree). Even if we choose not to utilize their DNS servers, by default all queries are in clear text and very easy to pick up.
The fact that now DNS functions over encrypted paths has certainly done a fair bit to reduce the ISP aspect, but we still need to trust someone (i.e. the recursive resolver we use) as they can see all queries.
Next step would be to use our own DNS server to handle all the queries (whether on the router or using some other tech on our networks, like pihole etc).
Not to delve to deep, the key point is, we need to decide whom we trust. Is it google, or 1.1.1.1 or quad9? Or, depending on location, folks like CIRA who run the Canadian Shield DNS servers with different levels of filtering.
And we're still at the Network layer. Haven't even gotten to Certificate Authorities. Then, it's time to tackle the applications, probably starting with OSs and moving onto browsers in short order. And the fact that many of us live in 5 eyes nations...
TL;DR: Security is tough because somewhere you need to trust people, organizations (governments?!), it's inevitable.
So, who do you trust?
Sorry for the long post.