To tackle the COVID-19 pandemic, massive efforts have been made in modeling COVID-19 transmission, diagnoses, interventions, pathological and influence analysis, etc. With the most comprehensive repository on COVID-19 research - some 160k WHO-collected global literature on COVID-19 produced since 2020, some critical question to ask include: What are the COVID-19 challenges? How do they address the challenges? Where are the significant gaps and opportunities in COVID-19 modeling?. Accordingly, informed by their statistics and a deep keyword-based similarity analysis of those references on COVID-19 modeling, this is the first to systemically summarize the disease, data and modeling challenges and the corresponding modeling progress and gaps. We come up with a transdisciplinary research landscape to summarize and match the business goals and tasks and their learning methods of COVID-19 modeling.